A Health Social Network Recommender System

Size: px
Start display at page:

Download "A Health Social Network Recommender System"

Transcription

1 A Health Social Network Recommender System Insu Song 1, Denise Dillon 2,TzeJuiGoh 3, and Min Sung 3 1,2 School of Business Information Technology, and Psychology James Cook University Australia {insu.song,denise.dillon}@jcu.edu.au 3 Institute of Mental Health (IMH), Singapore {Tze Jui Goh,Min SUNG}@imh.com.sg Abstract. People with chronic health conditions require support beyond normal health care systems. Social networking has shown great potential to provide the needed support. Because of the privacy and security issues of health information systems, it is often difficult to find patients who can support each other in the community. We propose a social-networking framework for patient care, in particular for parents of children with Autism Spectrum Disorders (ASD). In the framework, health service providers facilitate social links between parents using similarities of assessment reports without revealing sensitive information. A machine learning approach was developed to generate explanations of ASD assessments in order to assist clinicians in their assessment. The generated explanations are then used to measure similarities between assessments in order to recommend a community of related parents. For the first time, we report on the accuracy of social linking using an explanation-based similarity measure. Keywords: Social networking, health social network, health informatics, recommender system. 1 Introduction 1.1 Motivation Recently, social networking for health care has shown great potential to empower patient self-care. Examples include PatientsLikeMe 1 and the IBM Patient Empowerment System. These newly emerging patient-driven health care services facilitate information exchange and collaboration between patients and between patients and doctors. The services provided by health social networks include (a) emotional support and information sharing, (b) physician Q&As, and (c) self-tracking of a condition, its symptoms, treatment options and other biological information [16]. In this paper, we propose a social networking framework for parents of autistic children. Autism is characterized by a triad of impairments [18] in the areas of reciprocal social interaction, communication and repetitive and stereotyped behaviors. Asperger s Disorder is similar to Autism, but involves no deviance or delay in language development. Studies have demonstrated that early diagnosis can lead to better prognosis 1 ( D. Kinny et al. (Eds.): PRIMA 2011, LNAI 7047, pp , c Springer-Verlag Berlin Heidelberg 2011

2 362 I. Song et al. Fig. 1. (a) illustrates the overall process of generating textual explanations to classification results. (b) shows an example use of the explanation method, where a clinician make use of textual explanations to previous or current assessments, such as Autism. Mobile devices such as smart phones can be used to record interview questions and provide on the spot classification and explanations to provide more objective mental health assessments. for children with Autism Spectrum Disorders (ASD) [1]. However, most children get diagnosed only upon entering the school environment when the behavioral difficulties become more prominent. Hence, the implementation of early surveillance and screening is crucial to identify children at risk for ASD at an earlier stage [8,7,11]. Recently, machine learning techniques such as support vector machines (SVMs) have shown significant potential for supporting the practice of medicine and psychiatric classification [5]. The application of machine learning techniques in ASD diagnosis has significant merits because of the potential to provide early diagnosis and more standardized objective diagnosis. Conventionally, expert psychiatrists consciously and unconsciously analyze the language of their patients to make a clinical diagnosis using diagnostic classification schemas, such as the DSM IV [6] and ICD 10 [9]. Although the DSM-IV and ICD-10 guidelines are helpful to clinicians in the diagnostic process, the effectiveness of their utilization depends on the experience of the clinician [12]. ASD is usually a lifelong condition such that long-term treatment planning and supports from family and communities are essential. Social networking for health care may empower parents of autistic children to share information with other parents and more easily collaborate with doctors. In this paper we develop a method of facilitating social linking of parents by similarities of assessment reports of their children. Explanations of classification results of assessment reports are used to measure similarities of the assessment reports. The experiments describe a first attempt to generate explanations of why practicing psychiatrists would have diagnosed autism cases using a decompositional approach: learned SVM model parameters are analyzed to select informative features, and then sensitivity filtering is used to select some subsets of more relevant features. Figure 1 (a) illustrates an overview of our approach to generating explanations for psychological assessments using Support Vector Machines (SVMs). Explanation terms are extracted from assessment documents using both SVM models and classification results. Figure 1 (b) shows how our method can be used to provide explanation assisted assessment of autism and other mental health issues. For example, the explanationscan highlight the main issues that were used to differentiate the particular autism case from normal cases.

3 A Health Social Network Recommender System Background This work is based on social networking, text mining, text classification and, in particular, recommendation systems. The following section provides a brief overview of the core techniques, focusing on social networking, recommendation systems, support vector machines (SVMs), and the significance of generating human-comprehensible explanations from SVMs. Social Network and Recommendation System. A health social network is an online information service which facilitates information sharing between closely related members of a community. Also known as social media on the Internet, or Health 2.0, a health social network empowers patients and health service providers by promoting collaboration between patients, their caregivers, and clinicians [14]. At its basic level, a health social network provides emotional support by allowing patients to find others in similar health situations. They can also share information about conditions, symptoms and treatments [16]. Other services include physician Q&A, and self-tracking of condition, symptom, treatment and other biological information [16]. The self-supporting community is particularly important for lifelong conditions like autism. The main means of finding patients with similar health conditions are based on laborintensive methods such as searching the Internet, keywords in community titles and descriptions of other members in communities [15]. Over the years, many recommender systems and similarity measurement methods have been developed [3]. The approaches can be broadly classified into two categories: content matching based on available semantic information and a collaborative filtering approach based on overlapping membership of pairs of communities [15]. Our novel approach is based on semantic information of autism assessment reports. Support Vector Machines. Cortes and Vapnik [2] introduced support vector machines (SVMs) which are a novel approach to machine learning. SVMs are based on the structural risk minimization principle in order to overcome the overfitting problems. Support vector machines find the hypotheses out of the hypothesis space H of a learning system which approximately minimizes the bound on the actual error by controlling the empirical error using training samples and the complexity of the model using the VCdimension of H. SVMs are very universal learning systems [10]. In their basic form, SVMs learn maximal margin hyperplanes (linear threshold functions). A hyperplane can be defined by a weight vector w and a bias b: w x + b = 0 The corresponding threshold function for an input vector x is then given by: f (x)=sign(w x + b) However, it is possible learn polynomial classifiers, radial basis function (RBF) networks and three or more layered neural networks by mapping input data x to some other (possibly infinite dimensional) feature space φ(x) and using kernel functions K(x i,x j ) to obtain dot products, φ(x i ) φ(x j ), of feature data.

4 364 I. Song et al. Fig. 2. Parents of autistic children collaborate with hospital and the community to share experiences and learning to address their needs. The hospital engages in the community by providing links between parents having children with similar assessment results. Families are matched based on similarity of their explanation terms. Generating Explanations from SVMs. Much of the work that aims at providing an explanation capability to SVMs has focused on rule extraction techniques [4] following the footsteps of the earlier effort to obtain human-comprehensible rules from artificial neural networks (ANNs). One approach to classifying rule extraction methods is the translucency dimension which includes decompositional and pedagogical (or learning based) techniques as extremes [13]. The decompositional approach analyzes the internal representation of the ANN. In general, decompositional rule extraction techniques start with analyzing each individual neuron and their weight vectors to generate localized rules. Initially, the inputs and outputs of the neurons form antecedents and consequents of the rules, respectively. On the other hand, the strategy of the pedagogical approaches considers the trained ANN as a black box and aims at finding rules that map the ANN inputs directly to outputs [17]. For example, a decision tree can be generated from pairs of input and output values of the trained ANN. 1.3 Overview In the next section, we propose a social networking framework for parents of autistic children that utilizes mobile phone-based ubiquitous computing. The remainder of the paper summarizes experiments and their results: text classification, explanation generation for classification results, statistical analysis on the model parameters that are generated for autism diagnosis reports, and social linking using explanation similarities.

5 2 Parent Network Framework A Health Social Network Recommender System 365 We propose a parent social-network framework, where health service providers can actively engage in facilitating information sharing and social links between parents. Figure 2 shows how hospitals can interact with communities of parents. Parents who are concerned about their children can obtain preliminary assessment tools from hospitals via their mobile phones. They can fill in a standard assessment questionnaire to get a preliminary diagnosis and to obtain information on how to get help. Upon the first consultation with a clinician, the mobile agent on the parent can provide the clinician s agent completed questionnaires and other preliminary diagnosis results improving both the effectiveness and the efficiency of the clinician. The clinician can then, via the mobile agent of the parent, provide a treatment plan and tasks that parents can follow. The assessment report is then stored in the data mining server to generate explanations and parent-link information about other parents of children with similar diagnoses. Subsequent visits to the hospital will provide more refined treatment plans and information about communities who can share their experiences and should facilitate learning in order to meet the parents needs, such as emotional supports and clinical knowledge. 3 Experimental Evaluation 3.1 Methodology A preliminary study has been undertaken to generate explanations of autism diagnosis reports obtained from IMH (Institute of Mental Health, Singapore). Figure 3 illustrates our method of generating explanations. The autism diagnosis reports were obtained from mental health clinics and comprise of a total of 236 reports: 217 positive cases and 19 negativecases. A small part of the observation section of an autism assessment report is shown below (the sentences are paraphrased to protect the identity of patients): He had difficulties in responding to questions. He displayed difficulties in expressing himself and responded with only short incomplete sentences. He would respond with body gestures or single words when asked to elaborate on his responses. Often, he responded very slowly taking time to think before responding to questions. The autism text documents are represented as attribute-value vectors ( bag of words representation) where each distinct word corresponds to a feature whose value is the frequency of the word in the text sample. A text document is represented as a feature vector x =(x 1,.,x j,..,x L ) where x j is the j-th feature. Values were transformed with regard to the length of the sample. Function words were removed and stemming was performed on each extracted text. In summary, input vectors for machine learning consist of attributes (the words used in the sample) and values (the transformed frequency of the words). Outputs are autism versus normal, that is, binary decision tasks were learned. Clearly, the expressive power of the resulting explanations is limited by this bag of words representation.

6 366 I. Song et al. Fig. 3. Overview of the methodology and experiment of generating explanations to autism diagnosis result For LOO (leave-one-out) cross validation, 236 SVM models were generated using the linear kernel for the autism assessment data sets. Thus, each model is used to classify one document. An SVM model is defined by support vectors x i and associated parameters. The decision value of a text sample (represented as a feature vector x) is then obtained as follows: d(x)= α i y i K(x i,x)+b where x i are support vectors and x is the feature vector, α i are Lagrangian multipliers, and b is the offset. The antecedent of the rule of inference is then this: α i y i x i x + b 0 That is, if d(x) 0, the feature vector x is positive or else negative. We use this insight into the SVM models to define three types of explanations: 1. Explanation A comprising all the features contributing to the decision value d(x); 2. Explanation B comprising top-n contributing features that are sufficient to classify the features; 3. Explanation C comprising top-n contributing features that also have their sensitivity values d/ x j greater than a set threshold value τ.

7 A Health Social Network Recommender System 367 Fig. 4. (a) Relationship between contribution (deviation), sensitivity, and word ranks. Each point is a feature component that contributes to the decision of a feature. If a feature is a positive (negative) case, only the feature components having positive (negative) contributions are plotted. Rank 1 represents the most frequent term. (b) True-positive rate vs. false-positive rate of a linear support vector machine. Technical details on generating each explanation type are described in Section 4. This approach is clearly a decompositional approach: analysis on the model parameters to select informative components and selecting subsets of more relevant components. Figure 4a summarizes the significance of each type of explanations. It plots contribution, sensitivity, and word rank of all features of the autism data set. It shows that sample features with higher ranking orders (more frequent words) and higher sensitivity values tend to have larger contribution values. This suggests that features having higher sensitivity values and higher ranking order provide greater information in decision making than other features. It also shows that most of the large contributions are made by more frequent words (high rank words). 3.2 Results Support vector machines trained on the autism assessment data set achieved an accuracy of 90% and AUC (Area Under the Curve) of The corresponding ROC curve is shown in Figure 4 (b). The sensitivity values are adjusted manually to obtain a reasonable amount of terms for Explanation type C. Sample explanations of a positive autism diagnosis case are provided below (Explanation A samples too big to show here): 1. Explanation B: social (94 46), mother (53 18), brother (50 23), old (44 58), interest (39 27), game (28 24), describe (27 24), share (21 41), computer (18 31), family (18 33), resource (17 42), limit (16 36), information (16 62), create (13 28), Strategies (11 36),.., (omitted the rest).

8 368 I. Song et al. 2. Explanation C: social (94 46), old (44 58), interest (39 27), share (21 41), computer (18 31), family (18 33), resource (17 42), limit (16 36), inform (16 62), create (13 28), strategies (11 36),.., (omitted the rest). The numbers (d(x) i, d/ x i ) indicate relative contribution values d(x) i to the decision value d(x) and sensitivity d/ x i of the i-th term, respectively. Sensitivity-filtering (Explanation C) eliminates some of less sensitive terms (bold-faced terms) from Explanation B. Sample explanations of a negative autism diagnosis case are provided below: 1. Explanation B: average ( ), appropriate ( ), his (-84-18), attention (-62-84), during (-41-29), children (-32-40), indicated (-20-51), good (-19-57), age (-19-29), reason (-18-20), attempt (-16-26), regular (-14-12), in (-12-10), apparently (-11-26), mental (-10-22), well (-9-35),.., (omitted the rest). 2. Explanation C: average ( ), appropriate ( ), attention (-62-84), during (-41-29), children (-32-40), indicated (-20-51), good (-19-57), age (-19-29), attempt (-16-26), apparently (-11-26), well (-9-35),.., (omitted the rest). Negative cases have negative contribution and sensitivity values. 4 Generating Explanations from SVM Models In order to calculate the contribution values of each feature of a feature vector x, we use the centroid C of the population, which is estimated using the centroid C sv of the support vectors: C sv = 1 Nsv φ(x i ) where N sv is the number of support vectors. We can then calculate the deviation of a feature vector x from the estimate population centroid: D(x)=φ(x) C sv Suppose C sv is on the hyperplane: w φ(c sv ) b. Then, we can obtain the decision value d(x) using the deviation D(x): d(x) (w (φ(x) C sv )) = α i y i [K(x i,x) K(x i,c sv )] If K is the linear kernel, we can estimate the contribution of each j-th feature x j as follows: C sv, j = 1 Nsv x i, j d(x) j = α i y i x i, j (x j C sv, j ) Now, for a feature vector x, we can explain why a sample is positive (negative) by listing the feature elements that contribute to the decision value. That is, we can rank

9 A Health Social Network Recommender System 369 the features of a feature vector according to the amount of contributions made by the features. This is used as the basis of the explanation type A. We can also calculate the sum of all negative (positive) contributions and choose the top N positive (negative) contributions that are sufficient to push the decision value to positive (negative). This is used as the basis of the explanation type B. It can be shown that Explanation A, B, and C are consistent: the same features are not used to explain an opposite class. Consistency is one of the criteria for evaluating rule quality (Andrew et al. [13]). Other important criteria for evaluating rule quality are accuracy and fidelity. It can be shown that the accuracy of an SVM model is bounded by the accuracy of explanation terms. Furthermore, we can achieve a similar performance using only the explanation-terms as the vocabulary: explanation-terms can mimic the behavior of the SVM model from which the explanation terms are extracted. That is, the explanations display a high level of fidelity. This method can easily be extended to non-linear SVM models with convex decision boundaries. Applying the K-NN algorithm, N number of support vectors can be selected as an explanation reference point forming a centroid and a hyperplane in the input space. This new hyperplane is now a linear SVM model that can be used to generate explanations with regard to the selected support vectors. 4.1 Filtering Explanations with Sensitivity Training a support vector machine for a data set of interest generates a hyperplane, which can be used to obtain the distance of a feature vector to the hyperplane to classify. The distance is normal to the hyperplane and thus the importance of a feature can be measured as the rate of change of the distance with respect to the feature. This can be easily obtained for a linear classifier as follows: d(x) x j = α i y i x i, j where d(x) is the distance of feature x to the hyperplane, x j is the j-th component of the feature x, andx i, j is the j-th component of a support vector x i. As we can see from the above equation, the importance of the j-th component for the hyperplane is the sum of j-th component of the support vectors multiplied by the class label and the Lagrange multipliers. 5 Social Linking of Parents by Explanation Similarities The explanations generated provide relevancy of each feature to the particular classes. We can use this information to measure the relevancy of each part of the explanations to measure similarities between assessments. We use a semantic similarity measure between two terms based on a common sense database called ConceptNet 2. This measure then can be used to link parents having children with similar assessment results. 2 Used ConceptNet v2.1 from the Common Sense Computing Initiative at the MIT Media Lab (

10 370 I. Song et al. Fig. 5. Error rates of linking 18 autism assessment reports to top-k most similar assessment reports using top-n most contributing explanation terms We start with a simple approach to generating similarity measures. The method is scoring each explanation term in one assessment with each explanation term in another assessment. In this approach, the similarity between two assessments is given by determining contributions made by explanations terms and semantic relationships between terms. The similarity between two explanations A and B are defined as follows: s i, j = 1 A B u(i, j) d(a) i d(b) j i A j B {i} where u(i, j) is the semantic similarity function that measures how close the term i in explanation A is to the term j in an explanation B where i j, d(a) i and d(a) j are the amount of contributions of the features i and j, respectively. ConceptNet analogy space is used as the similarity function. Each semantic similarity between terms is the L1 similarity measure for social networks defined in [15]. That is, dot products of two vectors i and j in the analogy space, but weighted with contributions of terms. The parent link information is then generated by ranking assessments that are closed to an assessment and selecting N most similar assessments. To test the effectiveness of this method, similarities between 16 assessments were measured: 8 assessments with autism diagnosis and 8 assessments with negative autism diagnosis. The average of the similarity measure between assessments with same diagnosis results was 8.66 and the average of the similarity measures between assessments with different diagnosis results was The method of measuring similarities did not have information on class labels, but was able to distinguish positive cases from negative cases only using semantic similarity between the explanation terms. Figure 5 shows the error rates of linking parents to other parents with the same diagnosis results, where the error rate is defined as follows: Error Rate = Number of Incorrect Recommendations Number of Recommendations

11 A Health Social Network Recommender System 371 The parents were linked by selecting top-k most similar assessments using top-n most contributing explanation terms. It shows that when one parent was linked to 32% or less proportions of the total community, the error rate is less than 35%, and it consistently gets better as K decreases further. The average error rate matrix in Figure 5 (b) clearly indicates that a small number of key explanation terms can provide good link information and that the relevance information of explanation terms is useful. Using this similarity measure, we can recommend a parent p to a community in a set C of communities by selecting the community with the maximum average-similarity between the parent p and all parents p in a community c: R(p,C)=argmax c C 1 c p c s p,p 6 Discussions and Future Work This is the first report of a novel approach in providing social network-based health care services to families with ASD children. This is also the first report on the accuracy of social-linking using an explanation-based similarity measure. We showed that a semantic similarity measure between explanations of assessments has great potential for discovering close social communities of parents who can support each other for lifelong conditions like autism. It can also be used to find all potential hidden communities of patients and parents on the Internet. There is massive potential of incorporating these sophisticated information extraction technologies in social networking more generally. Long term disabilities likes ASD pose a significant burden for families, and thus it is essential that health care services actively participate in communities to support patients families. The community suggestion method facilitates social linking between parents with similar assessment reports to make information obtained through social networking more relevant and useful. The approach of extracting some piece of knowledge using machine learning in explaining psychiatric assessments has the potential to provide early diagnosis and more standardized assessments, and to improve the usability of machine learning techniques in the medical and security domains. This approach of extracting explanations using some form of analysis on machine learning and associated parameters can be further expanded by using alternative feature representations of text data sets, such as concept terms or semantic terms. References 1. Charman, T., Baird, G.: Practitioner review: Diagnosis of autism spectrum disorders in 2- and 3-year-old children. Journal of Child Psychology and Psychiatry 43(3), (2002) 2. Cortes, C., Vapnik, V.: Support-vector networks. Machine Learning 20(3), (1995) 3. Deshpande, M., Karypis, G.: Item-based top-n recommendation algorithms. ACM Trans. Inf. Syst. 22(1), (2004) 4. Diederich, J.: Rule extraction from support vector machines: An introduction. In: Rule Extraction from Support Vector Machines. SCI, vol. 80, pp Springer, Heidelberg (2008)

12 372 I. Song et al. 5. Diederich, J., Al-Ajmi, A., Yellowlees, P.: Ex-ray: Data mining and mental health. Applied Soft Computing 7(3), (2007) 6. DSMIV: Diagnostic and Statistical Manual of Mental Disorders, Text Revision, 4th edn. American Psychiatric Association (2000) 7. Filipek, P.A., Accardo, P.J., Baranek, G.T., Cook, J.E.H., Dawson, G., Gordon, B., et al.: The screening and diagnosis of autistic spectrum disorders. Journal of Autism and Developmental Disorders 29(6), (1999) 8. Filipek, P.A., Accardo, P.J., Ashwal, S., Baranek, G.T., Cook Jr., E.H., Dawson, G., et al.: Practice parameter: Screening and diagnosis of autism. report of the quality standards subcommittee of the american academy of neurology and the child neurology society. Neurology 55(3), (2000) 9. ICD10: International Statistical Classification of Disease and Related Health. World Health Organization, Geneva (1992) 10. Joachims, T.: Making large-scale support vector machine learning practical, pp (1999) 11. Johnson, C.P., Myers, S.M.: The Council on Children with Disabilities: Identification and evaluation of children with autism spectrum disorders. Pediatrics 120, (2007) 12. Klin, A., Lang, J., Cicchetti, D.V., Volkmar, F.: Brief report: Interrater reliability of clinical diagnosis and dsm-iv criteria for autistic disorder: Results of the dsm-iv autism field trial. Journal of Autism and Developmental Disorders 30(2), (2000) 13. Robert Andrews, J.D., Tickle, A.B.: Survey and critique of techniques for extracting rules from trained artificial neural networks. Knowledge-Based Systems 8(6), (1995) 14. Sarasohn-Kahn, J.: The wisdom of patients: Health care meets online social media. California HealthCare Foundation iheath Reports (April 2008), Spertus, E., Sahami, M., Buyukkokten, O.: Evaluating similarity measures: a large-scale study in the orkut social network. In: Proceedings of the Eleventh ACM SIGKDD International Conference on Knowledge Discovery in Data Mining, KDD 2005, pp ACM, New York (2005) 16. Swan, M.: Emerging patient-driven health care models: an examination of health social networks, consumer personalized medicine and quantified self-tracking. International Journal of Environmental Research and Public Health 6(2), (2009) 17. Tickle, A., Andrews, R., Golea, M., Diederich, J.: The truth will come to light: Directions and challenges in extracting the knowledge embedded within trained artificial neural networks. IEEE Transactions on Neural Networks 9(6), (1998) 18. Wing, L., Gould, J.: Severe impairments of social interaction and associated abnormalities in children: Epidemiology and classification. Journal of Autism and Developmental Disorders. 9(1), (1979)

First Concerns. Wh at if I (o r t h e pa r e n t s) h av e c o n c e r n s a b o u t a pat i e n t? 10 Toolkit for Medical Professionals

First Concerns. Wh at if I (o r t h e pa r e n t s) h av e c o n c e r n s a b o u t a pat i e n t? 10 Toolkit for Medical Professionals 10 Toolkit for Medical Professionals 1 First Concerns Wh at if I (o r t h e pa r e n t s) h av e c o n c e r n s a b o u t a pat i e n t? Remember that the AAP guidelines now indicate that all children

More information

Studies in Computational Intelligence

Studies in Computational Intelligence Studies in Computational Intelligence Volume 491 Series Editor Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland e-mail: kacprzyk@ibspan.waw.pl For further volumes: http://www.springer.com/series/7092

More information

Data Mining in Bioinformatics Day 4: Text Mining

Data Mining in Bioinformatics Day 4: Text Mining Data Mining in Bioinformatics Day 4: Text Mining Karsten Borgwardt February 25 to March 10 Bioinformatics Group MPIs Tübingen Karsten Borgwardt: Data Mining in Bioinformatics, Page 1 What is text mining?

More information

A Comparison of Collaborative Filtering Methods for Medication Reconciliation

A Comparison of Collaborative Filtering Methods for Medication Reconciliation A Comparison of Collaborative Filtering Methods for Medication Reconciliation Huanian Zheng, Rema Padman, Daniel B. Neill The H. John Heinz III College, Carnegie Mellon University, Pittsburgh, PA, 15213,

More information

Starting Points. Starting Points. Autism Screening and Resources for the Practitioner. The Importance of Screening

Starting Points. Starting Points. Autism Screening and Resources for the Practitioner. The Importance of Screening Autism Screening and Resources for the Practitioner Ruth Aspy, Ph.D., and Barry G. Grossman, Ph.D. The Ziggurat Group, PA www.texasautism.com Keynote Presented for the Greater Texas Chapter National Association

More information

UNIVERSITY of PENNSYLVANIA CIS 520: Machine Learning Final, Fall 2014

UNIVERSITY of PENNSYLVANIA CIS 520: Machine Learning Final, Fall 2014 UNIVERSITY of PENNSYLVANIA CIS 520: Machine Learning Final, Fall 2014 Exam policy: This exam allows two one-page, two-sided cheat sheets (i.e. 4 sides); No other materials. Time: 2 hours. Be sure to write

More information

Mammogram Analysis: Tumor Classification

Mammogram Analysis: Tumor Classification Mammogram Analysis: Tumor Classification Term Project Report Geethapriya Raghavan geeragh@mail.utexas.edu EE 381K - Multidimensional Digital Signal Processing Spring 2005 Abstract Breast cancer is the

More information

Data mining for Obstructive Sleep Apnea Detection. 18 October 2017 Konstantinos Nikolaidis

Data mining for Obstructive Sleep Apnea Detection. 18 October 2017 Konstantinos Nikolaidis Data mining for Obstructive Sleep Apnea Detection 18 October 2017 Konstantinos Nikolaidis Introduction: What is Obstructive Sleep Apnea? Obstructive Sleep Apnea (OSA) is a relatively common sleep disorder

More information

Mammogram Analysis: Tumor Classification

Mammogram Analysis: Tumor Classification Mammogram Analysis: Tumor Classification Literature Survey Report Geethapriya Raghavan geeragh@mail.utexas.edu EE 381K - Multidimensional Digital Signal Processing Spring 2005 Abstract Breast cancer is

More information

A Learning Method of Directly Optimizing Classifier Performance at Local Operating Range

A Learning Method of Directly Optimizing Classifier Performance at Local Operating Range A Learning Method of Directly Optimizing Classifier Performance at Local Operating Range Lae-Jeong Park and Jung-Ho Moon Department of Electrical Engineering, Kangnung National University Kangnung, Gangwon-Do,

More information

Introduction to Machine Learning. Katherine Heller Deep Learning Summer School 2018

Introduction to Machine Learning. Katherine Heller Deep Learning Summer School 2018 Introduction to Machine Learning Katherine Heller Deep Learning Summer School 2018 Outline Kinds of machine learning Linear regression Regularization Bayesian methods Logistic Regression Why we do this

More information

A Vision-based Affective Computing System. Jieyu Zhao Ningbo University, China

A Vision-based Affective Computing System. Jieyu Zhao Ningbo University, China A Vision-based Affective Computing System Jieyu Zhao Ningbo University, China Outline Affective Computing A Dynamic 3D Morphable Model Facial Expression Recognition Probabilistic Graphical Models Some

More information

An Improved Algorithm To Predict Recurrence Of Breast Cancer

An Improved Algorithm To Predict Recurrence Of Breast Cancer An Improved Algorithm To Predict Recurrence Of Breast Cancer Umang Agrawal 1, Ass. Prof. Ishan K Rajani 2 1 M.E Computer Engineer, Silver Oak College of Engineering & Technology, Gujarat, India. 2 Assistant

More information

Classification. Methods Course: Gene Expression Data Analysis -Day Five. Rainer Spang

Classification. Methods Course: Gene Expression Data Analysis -Day Five. Rainer Spang Classification Methods Course: Gene Expression Data Analysis -Day Five Rainer Spang Ms. Smith DNA Chip of Ms. Smith Expression profile of Ms. Smith Ms. Smith 30.000 properties of Ms. Smith The expression

More information

Variable Features Selection for Classification of Medical Data using SVM

Variable Features Selection for Classification of Medical Data using SVM Variable Features Selection for Classification of Medical Data using SVM Monika Lamba USICT, GGSIPU, Delhi, India ABSTRACT: The parameters selection in support vector machines (SVM), with regards to accuracy

More information

Technical Specifications

Technical Specifications Technical Specifications In order to provide summary information across a set of exercises, all tests must employ some form of scoring models. The most familiar of these scoring models is the one typically

More information

Contents. Just Classifier? Rules. Rules: example. Classification Rule Generation for Bioinformatics. Rule Extraction from a trained network

Contents. Just Classifier? Rules. Rules: example. Classification Rule Generation for Bioinformatics. Rule Extraction from a trained network Contents Classification Rule Generation for Bioinformatics Hyeoncheol Kim Rule Extraction from Neural Networks Algorithm Ex] Promoter Domain Hybrid Model of Knowledge and Learning Knowledge refinement

More information

Supplementary Online Content 2

Supplementary Online Content 2 Supplementary Online Content 2 Bieleninik Ł, Geretsegger M, Mössler K, et al; TIME-A Study Team. Effects of improvisational music therapy vs enhanced standard care on symptom severity among children with

More information

Autism/Pervasive Developmental Disorders Update. Kimberly Macferran, MD Pediatric Subspecialty for the Primary Care Provider December 2, 2011

Autism/Pervasive Developmental Disorders Update. Kimberly Macferran, MD Pediatric Subspecialty for the Primary Care Provider December 2, 2011 Autism/Pervasive Developmental Disorders Update Kimberly Macferran, MD Pediatric Subspecialty for the Primary Care Provider December 2, 2011 Overview Diagnostic criteria for autism spectrum disorders Screening/referral

More information

Background on the issue Previous study with adolescents and adults: Current NIH R03 study examining ADI-R for Spanish speaking Latinos

Background on the issue Previous study with adolescents and adults: Current NIH R03 study examining ADI-R for Spanish speaking Latinos Sandy Magaña Background on the issue Previous study with adolescents and adults: brief description of study examining comparison between whites and Latinos in on the ADI-R Current NIH R03 study examining

More information

38 Int'l Conf. Bioinformatics and Computational Biology BIOCOMP'16

38 Int'l Conf. Bioinformatics and Computational Biology BIOCOMP'16 38 Int'l Conf. Bioinformatics and Computational Biology BIOCOMP'16 PGAR: ASD Candidate Gene Prioritization System Using Expression Patterns Steven Cogill and Liangjiang Wang Department of Genetics and

More information

Zainab M. AlQenaei. Dissertation Defense University of Colorado at Boulder Leeds School of Business Operations and Information Management Division

Zainab M. AlQenaei. Dissertation Defense University of Colorado at Boulder Leeds School of Business Operations and Information Management Division An Investigation of the Relationship between Consumer Mental Health Recovery Indicators and Clinicians Reports Using Multivariate Analyses of the Singular Value Decomposition of a Textual Corpus Zainab

More information

Experimental Research in HCI. Alma Leora Culén University of Oslo, Department of Informatics, Design

Experimental Research in HCI. Alma Leora Culén University of Oslo, Department of Informatics, Design Experimental Research in HCI Alma Leora Culén University of Oslo, Department of Informatics, Design almira@ifi.uio.no INF2260/4060 1 Oslo, 15/09/16 Review Method Methodology Research methods are simply

More information

6/5/2018 SYLVIA J. ACOSTA, PHD

6/5/2018 SYLVIA J. ACOSTA, PHD SYLVIA J. ACOSTA, PHD ASSOCIATE PROFESSOR SUMMER INSTITUTE JUNE 1 Introduction to Autism Spectrum Disorder (ASD) for Educators JUNE 15, 2018 2 Objectives Participants will: Identify the 2 diagnostic categories

More information

The Nuts and Bolts of Diagnosing Autism Spectrum Disorders In Young Children. Overview

The Nuts and Bolts of Diagnosing Autism Spectrum Disorders In Young Children. Overview The Nuts and Bolts of Diagnosing Autism Spectrum Disorders In Young Children Jessica Greenson, Ph.D. Autism Center University of Washington Overview Diagnostic Criteria Current: Diagnostic & Statistical

More information

A Review on Sleep Apnea Detection from ECG Signal

A Review on Sleep Apnea Detection from ECG Signal A Review on Sleep Apnea Detection from ECG Signal Soumya Gopal 1, Aswathy Devi T. 2 1 M.Tech Signal Processing Student, Department of ECE, LBSITW, Kerala, India 2 Assistant Professor, Department of ECE,

More information

SUPPLEMENTARY INFORMATION. Table 1 Patient characteristics Preoperative. language testing

SUPPLEMENTARY INFORMATION. Table 1 Patient characteristics Preoperative. language testing Categorical Speech Representation in the Human Superior Temporal Gyrus Edward F. Chang, Jochem W. Rieger, Keith D. Johnson, Mitchel S. Berger, Nicholas M. Barbaro, Robert T. Knight SUPPLEMENTARY INFORMATION

More information

Getting Started with AAC

Getting Started with AAC Getting Started with AAC P A R E N T G U I D E Many children have medical conditions that impact their ability to speak and learn language. But thanks to augmentative and alternative communication (AAC),

More information

Collaborative, evidence based understanding of students with ASD

Collaborative, evidence based understanding of students with ASD Collaborative, evidence based understanding of students with ASD Rebecca Sutherland Speech Pathologist Child Development Unit Children s Hospital at Westmead Positive Partnerships www.positivepartnerships.com.au

More information

Winner s Report: KDD CUP Breast Cancer Identification

Winner s Report: KDD CUP Breast Cancer Identification Winner s Report: KDD CUP Breast Cancer Identification ABSTRACT Claudia Perlich, Prem Melville, Yan Liu, Grzegorz Świrszcz, Richard Lawrence IBM T.J. Watson Research Center Yorktown Heights, NY 10598 {perlich,pmelvil,liuya}@us.ibm.com

More information

Lung Cancer Diagnosis from CT Images Using Fuzzy Inference System

Lung Cancer Diagnosis from CT Images Using Fuzzy Inference System Lung Cancer Diagnosis from CT Images Using Fuzzy Inference System T.Manikandan 1, Dr. N. Bharathi 2 1 Associate Professor, Rajalakshmi Engineering College, Chennai-602 105 2 Professor, Velammal Engineering

More information

Chapter 12 Conclusions and Outlook

Chapter 12 Conclusions and Outlook Chapter 12 Conclusions and Outlook In this book research in clinical text mining from the early days in 1970 up to now (2017) has been compiled. This book provided information on paper based patient record

More information

Predicting Breast Cancer Recurrence Using Machine Learning Techniques

Predicting Breast Cancer Recurrence Using Machine Learning Techniques Predicting Breast Cancer Recurrence Using Machine Learning Techniques Umesh D R Department of Computer Science & Engineering PESCE, Mandya, Karnataka, India Dr. B Ramachandra Department of Electrical and

More information

Efficient Classification of Cancer using Support Vector Machines and Modified Extreme Learning Machine based on Analysis of Variance Features

Efficient Classification of Cancer using Support Vector Machines and Modified Extreme Learning Machine based on Analysis of Variance Features American Journal of Applied Sciences 8 (12): 1295-1301, 2011 ISSN 1546-9239 2011 Science Publications Efficient Classification of Cancer using Support Vector Machines and Modified Extreme Learning Machine

More information

Brain Tumor segmentation and classification using Fcm and support vector machine

Brain Tumor segmentation and classification using Fcm and support vector machine Brain Tumor segmentation and classification using Fcm and support vector machine Gaurav Gupta 1, Vinay singh 2 1 PG student,m.tech Electronics and Communication,Department of Electronics, Galgotia College

More information

Stepwise Knowledge Acquisition in a Fuzzy Knowledge Representation Framework

Stepwise Knowledge Acquisition in a Fuzzy Knowledge Representation Framework Stepwise Knowledge Acquisition in a Fuzzy Knowledge Representation Framework Thomas E. Rothenfluh 1, Karl Bögl 2, and Klaus-Peter Adlassnig 2 1 Department of Psychology University of Zurich, Zürichbergstraße

More information

8/5/2018. Parent Implemented Interventions for Infants & Toddlers at risk for or with ASD

8/5/2018. Parent Implemented Interventions for Infants & Toddlers at risk for or with ASD Els Center of Excellence 18370 Limestone Creek Road Jupiter, FL 33458 Phone: 561 320 9520 Parent Implemented Interventions for Infants & Toddlers at risk for or with ASD Erin Brooker Lozott, M.S. CCC SLP

More information

What is Autism? Laura Ferguson, M.Ed., BCBA.

What is Autism? Laura Ferguson, M.Ed., BCBA. What is Autism? Laura Ferguson, M.Ed., BCBA. What is Autism? ) Autism is a complex developmental disability that has a neurological basis that causes impairments in social interactions, communication,

More information

Gene Selection for Tumor Classification Using Microarray Gene Expression Data

Gene Selection for Tumor Classification Using Microarray Gene Expression Data Gene Selection for Tumor Classification Using Microarray Gene Expression Data K. Yendrapalli, R. Basnet, S. Mukkamala, A. H. Sung Department of Computer Science New Mexico Institute of Mining and Technology

More information

THE data used in this project is provided. SEIZURE forecasting systems hold promise. Seizure Prediction from Intracranial EEG Recordings

THE data used in this project is provided. SEIZURE forecasting systems hold promise. Seizure Prediction from Intracranial EEG Recordings 1 Seizure Prediction from Intracranial EEG Recordings Alex Fu, Spencer Gibbs, and Yuqi Liu 1 INTRODUCTION SEIZURE forecasting systems hold promise for improving the quality of life for patients with epilepsy.

More information

Application of Artificial Neural Networks in Classification of Autism Diagnosis Based on Gene Expression Signatures

Application of Artificial Neural Networks in Classification of Autism Diagnosis Based on Gene Expression Signatures Application of Artificial Neural Networks in Classification of Autism Diagnosis Based on Gene Expression Signatures 1 2 3 4 5 Kathleen T Quach Department of Neuroscience University of California, San Diego

More information

The Action Is In the Interaction

The Action Is In the Interaction Evidence Base for the DIRFloortime Approach Diane Cullinane, M.D. 02-2015 DIR/Floortime is a way of relating to a child in which we recognize and respect the emotional experience of the child, shown in

More information

Efficacy of the Extended Principal Orthogonal Decomposition Method on DNA Microarray Data in Cancer Detection

Efficacy of the Extended Principal Orthogonal Decomposition Method on DNA Microarray Data in Cancer Detection 202 4th International onference on Bioinformatics and Biomedical Technology IPBEE vol.29 (202) (202) IASIT Press, Singapore Efficacy of the Extended Principal Orthogonal Decomposition on DA Microarray

More information

Deconstructing the DSM-5 By Jason H. King

Deconstructing the DSM-5 By Jason H. King Deconstructing the DSM-5 By Jason H. King Assessment and diagnosis of autism spectrum disorder For this month s topic, I am excited to share my recent experience using the fifth edition of the Diagnostic

More information

Evaluating the Behavioral and Developmental Interventions for Autism Spectrum Disorder

Evaluating the Behavioral and Developmental Interventions for Autism Spectrum Disorder International Journal of Information Sciences and Application. ISSN 0974-2255 Volume 6, Number 1 (2014), pp. 1-10 International Research Publication House http://www.irphouse.com Evaluating the Behavioral

More information

Autism Spectrum Disorders: An update on research and clinical practices for SLPs

Autism Spectrum Disorders: An update on research and clinical practices for SLPs DSM-IV to DSM-5: Primary Changes Autism Spectrum Disorders: An update on research and clinical practices for SLPs Laurie Swineford, PhD CCC-SLP Washington State University DSM-IV Previously we used the

More information

Improved Intelligent Classification Technique Based On Support Vector Machines

Improved Intelligent Classification Technique Based On Support Vector Machines Improved Intelligent Classification Technique Based On Support Vector Machines V.Vani Asst.Professor,Department of Computer Science,JJ College of Arts and Science,Pudukkottai. Abstract:An abnormal growth

More information

Empirical Mode Decomposition based Feature Extraction Method for the Classification of EEG Signal

Empirical Mode Decomposition based Feature Extraction Method for the Classification of EEG Signal Empirical Mode Decomposition based Feature Extraction Method for the Classification of EEG Signal Anant kulkarni MTech Communication Engineering Vellore Institute of Technology Chennai, India anant8778@gmail.com

More information

Case Studies of Signed Networks

Case Studies of Signed Networks Case Studies of Signed Networks Christopher Wang December 10, 2014 Abstract Many studies on signed social networks focus on predicting the different relationships between users. However this prediction

More information

Assessing Functional Neural Connectivity as an Indicator of Cognitive Performance *

Assessing Functional Neural Connectivity as an Indicator of Cognitive Performance * Assessing Functional Neural Connectivity as an Indicator of Cognitive Performance * Brian S. Helfer 1, James R. Williamson 1, Benjamin A. Miller 1, Joseph Perricone 1, Thomas F. Quatieri 1 MIT Lincoln

More information

Overview. Clinical Features

Overview. Clinical Features Jessica Greenson, Ph.D. Autism Center University of Washington Clinical Features Overview Diagnostic & Statistical Manual IV (DSM IV) Prevalence Course of Onset Etiology Early Recognition Early Recognition

More information

Predicting Breast Cancer Survivability Rates

Predicting Breast Cancer Survivability Rates Predicting Breast Cancer Survivability Rates For data collected from Saudi Arabia Registries Ghofran Othoum 1 and Wadee Al-Halabi 2 1 Computer Science, Effat University, Jeddah, Saudi Arabia 2 Computer

More information

Nature Neuroscience: doi: /nn Supplementary Figure 1. Behavioral training.

Nature Neuroscience: doi: /nn Supplementary Figure 1. Behavioral training. Supplementary Figure 1 Behavioral training. a, Mazes used for behavioral training. Asterisks indicate reward location. Only some example mazes are shown (for example, right choice and not left choice maze

More information

A micropower support vector machine based seizure detection architecture for embedded medical devices

A micropower support vector machine based seizure detection architecture for embedded medical devices A micropower support vector machine based seizure detection architecture for embedded medical devices The MIT Faculty has made this article openly available. Please share how this access benefits you.

More information

Identifying students with autism spectrum disorders: A review of selected screening tools

Identifying students with autism spectrum disorders: A review of selected screening tools Nova Southeastern University From the SelectedWorks of Lee A Wilkinson, PhD 2011 Identifying students with autism spectrum disorders: A review of selected screening tools Lee A Wilkinson Available at:

More information

Empirical function attribute construction in classification learning

Empirical function attribute construction in classification learning Pre-publication draft of a paper which appeared in the Proceedings of the Seventh Australian Joint Conference on Artificial Intelligence (AI'94), pages 29-36. Singapore: World Scientific Empirical function

More information

COMPARISON BETWEEN GMM-SVM SEQUENCE KERNEL AND GMM: APPLICATION TO SPEECH EMOTION RECOGNITION

COMPARISON BETWEEN GMM-SVM SEQUENCE KERNEL AND GMM: APPLICATION TO SPEECH EMOTION RECOGNITION Journal of Engineering Science and Technology Vol. 11, No. 9 (2016) 1221-1233 School of Engineering, Taylor s University COMPARISON BETWEEN GMM-SVM SEQUENCE KERNEL AND GMM: APPLICATION TO SPEECH EMOTION

More information

Gender Based Emotion Recognition using Speech Signals: A Review

Gender Based Emotion Recognition using Speech Signals: A Review 50 Gender Based Emotion Recognition using Speech Signals: A Review Parvinder Kaur 1, Mandeep Kaur 2 1 Department of Electronics and Communication Engineering, Punjabi University, Patiala, India 2 Department

More information

Design of Multi-Class Classifier for Prediction of Diabetes using Linear Support Vector Machine

Design of Multi-Class Classifier for Prediction of Diabetes using Linear Support Vector Machine Design of Multi-Class Classifier for Prediction of Diabetes using Linear Support Vector Machine Akshay Joshi Anum Khan Omkar Kulkarni Department of Computer Engineering Department of Computer Engineering

More information

Large-Scale Statistical Modelling via Machine Learning Classifiers

Large-Scale Statistical Modelling via Machine Learning Classifiers J. Stat. Appl. Pro. 2, No. 3, 203-222 (2013) 203 Journal of Statistics Applications & Probability An International Journal http://dx.doi.org/10.12785/jsap/020303 Large-Scale Statistical Modelling via Machine

More information

IJESRT. Scientific Journal Impact Factor: (ISRA), Impact Factor: 1.852

IJESRT. Scientific Journal Impact Factor: (ISRA), Impact Factor: 1.852 IJESRT INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH TECHNOLOGY Performance Analysis of Brain MRI Using Multiple Method Shroti Paliwal *, Prof. Sanjay Chouhan * Department of Electronics & Communication

More information

arxiv: v1 [cs.lg] 4 Feb 2019

arxiv: v1 [cs.lg] 4 Feb 2019 Machine Learning for Seizure Type Classification: Setting the benchmark Subhrajit Roy [000 0002 6072 5500], Umar Asif [0000 0001 5209 7084], Jianbin Tang [0000 0001 5440 0796], and Stefan Harrer [0000

More information

Early Autism Detection Screening and Referral. What is Autism? ASD Epidemiology. ASD Basic Facts 10/10/2010. Early Autism Detection and Referral

Early Autism Detection Screening and Referral. What is Autism? ASD Epidemiology. ASD Basic Facts 10/10/2010. Early Autism Detection and Referral Early Autism Detection and Referral Early Autism Detection Screening and Referral Learning Objectives: Define autistic spectrum disorders, their epidemiology and etiology; Recognize the earliest signs

More information

Comparison of Clinic & Home Observations of Social Communication Red Flags in Toddlers with ASD

Comparison of Clinic & Home Observations of Social Communication Red Flags in Toddlers with ASD Comparison of Clinic & Home Observations of Social Communication Red Flags in Toddlers with ASD David McCoy, Ph.D. California State University, Chico Sheri Stronach, University of Minnesota Juliann Woods

More information

Education Options for Children with Autism

Education Options for Children with Autism Empowering children with Autism and their families through knowledge and support Education Options for Children with Autism Starting school is a major milestone in a child s life, and a big step for all

More information

Abstract. Author. Costanza Colombi. Keywords: Autism Spectrum Disorder (ASD), Early Intervention, Challenges

Abstract. Author. Costanza Colombi. Keywords: Autism Spectrum Disorder (ASD), Early Intervention, Challenges Author Costanza Colombi ccolombi@umich.edu Abstract Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder that involves global impairments in social skills and in verbal and non-verbal communication,

More information

ABSTRACT I. INTRODUCTION. Mohd Thousif Ahemad TSKC Faculty Nagarjuna Govt. College(A) Nalgonda, Telangana, India

ABSTRACT I. INTRODUCTION. Mohd Thousif Ahemad TSKC Faculty Nagarjuna Govt. College(A) Nalgonda, Telangana, India International Journal of Scientific Research in Computer Science, Engineering and Information Technology 2018 IJSRCSEIT Volume 3 Issue 1 ISSN : 2456-3307 Data Mining Techniques to Predict Cancer Diseases

More information

The use of Autism Mental Status Exam in an Italian sample. A brief report

The use of Autism Mental Status Exam in an Italian sample. A brief report Life Span and Disability XX, 1 (2017), 93-103 The use of Autism Mental Status Exam in an Italian sample. A brief report Marinella Zingale 1, Simonetta Panerai 2, Serafino Buono 3, Grazia Trubia 4, Maurizio

More information

10/18/2016. Vineland Adaptive Behavior Scales, Third Edition 1. Meet Dr. Saulnier. Bio. Celine A. Saulnier, PhD Vineland-3 Author

10/18/2016. Vineland Adaptive Behavior Scales, Third Edition 1. Meet Dr. Saulnier. Bio. Celine A. Saulnier, PhD Vineland-3 Author Vineland Adaptive Behavior Scales, Third Edition Celine A. Saulnier, PhD Vineland-3 Author Director of Research Operations at the Marcus Autism Center & Associate Professor in the Department of Pediatrics

More information

Error Detection based on neural signals

Error Detection based on neural signals Error Detection based on neural signals Nir Even- Chen and Igor Berman, Electrical Engineering, Stanford Introduction Brain computer interface (BCI) is a direct communication pathway between the brain

More information

This is the accepted version of this article. To be published as : This is the author version published as:

This is the accepted version of this article. To be published as : This is the author version published as: QUT Digital Repository: http://eprints.qut.edu.au/ This is the author version published as: This is the accepted version of this article. To be published as : This is the author version published as: Chew,

More information

Machine Learning! Robert Stengel! Robotics and Intelligent Systems MAE 345,! Princeton University, 2017

Machine Learning! Robert Stengel! Robotics and Intelligent Systems MAE 345,! Princeton University, 2017 Machine Learning! Robert Stengel! Robotics and Intelligent Systems MAE 345,! Princeton University, 2017 A.K.A. Artificial Intelligence Unsupervised learning! Cluster analysis Patterns, Clumps, and Joining

More information

Predicting Sleep Using Consumer Wearable Sensing Devices

Predicting Sleep Using Consumer Wearable Sensing Devices Predicting Sleep Using Consumer Wearable Sensing Devices Miguel A. Garcia Department of Computer Science Stanford University Palo Alto, California miguel16@stanford.edu 1 Introduction In contrast to the

More information

The Open Access Institutional Repository at Robert Gordon University

The Open Access Institutional Repository at Robert Gordon University OpenAIR@RGU The Open Access Institutional Repository at Robert Gordon University http://openair.rgu.ac.uk This is an author produced version of a paper published in Intelligent Data Engineering and Automated

More information

Brief Report: Interrater Reliability of Clinical Diagnosis and DSM-IV Criteria for Autistic Disorder: Results of the DSM-IV Autism Field Trial

Brief Report: Interrater Reliability of Clinical Diagnosis and DSM-IV Criteria for Autistic Disorder: Results of the DSM-IV Autism Field Trial Journal of Autism and Developmental Disorders, Vol. 30, No. 2, 2000 Brief Report: Interrater Reliability of Clinical Diagnosis and DSM-IV Criteria for Autistic Disorder: Results of the DSM-IV Autism Field

More information

7.1 Grading Diabetic Retinopathy

7.1 Grading Diabetic Retinopathy Chapter 7 DIABETIC RETINOPATHYGRADING -------------------------------------------------------------------------------------------------------------------------------------- A consistent approach to the

More information

Exploratory Quantitative Contrast Set Mining: A Discretization Approach

Exploratory Quantitative Contrast Set Mining: A Discretization Approach Exploratory Quantitative Contrast Set Mining: A Discretization Approach Mondelle Simeon and Robert J. Hilderman Department of Computer Science University of Regina Regina, Saskatchewan, Canada S4S 0A2

More information

Essential behaviours for the diagnosis of DSM-5 Autism Spectrum Disorder

Essential behaviours for the diagnosis of DSM-5 Autism Spectrum Disorder Essential behaviours for the diagnosis of DSM-5 Autism Spectrum Disorder Dr Sarah Carrington Wales Autism Research Centre, Cardiff University Australasian Society for Autism Research 4 th December, 2014

More information

Chapter Three BRIDGE TO THE PSYCHOPATHOLOGIES

Chapter Three BRIDGE TO THE PSYCHOPATHOLOGIES Chapter Three BRIDGE TO THE PSYCHOPATHOLOGIES Developmental Psychopathology: From Infancy through Adolescence, 5 th edition By Charles Wenar and Patricia Kerig When do behaviors or issues become pathologies?

More information

Cover Page. The handle holds various files of this Leiden University dissertation.

Cover Page. The handle   holds various files of this Leiden University dissertation. Cover Page The handle http://hdl.handle.net/1887/19149 holds various files of this Leiden University dissertation. Author: Maljaars, Janne Pieternella Wilhelmina Title: Communication problems in children

More information

An Autism Primer for the PCP: What to Expect, When to Refer

An Autism Primer for the PCP: What to Expect, When to Refer An Autism Primer for the PCP: What to Expect, When to Refer Webinar November 9, 2016 John P. Pelegano MD Chief of Pediatrics Hospital for Special Care Disclosures None I will not be discussing any treatments,

More information

Classification of Honest and Deceitful Memory in an fmri Paradigm CS 229 Final Project Tyler Boyd Meredith

Classification of Honest and Deceitful Memory in an fmri Paradigm CS 229 Final Project Tyler Boyd Meredith 12/14/12 Classification of Honest and Deceitful Memory in an fmri Paradigm CS 229 Final Project Tyler Boyd Meredith Introduction Background and Motivation In the past decade, it has become popular to use

More information

Validating the Visual Saliency Model

Validating the Visual Saliency Model Validating the Visual Saliency Model Ali Alsam and Puneet Sharma Department of Informatics & e-learning (AITeL), Sør-Trøndelag University College (HiST), Trondheim, Norway er.puneetsharma@gmail.com Abstract.

More information

A Fuzzy Improved Neural based Soft Computing Approach for Pest Disease Prediction

A Fuzzy Improved Neural based Soft Computing Approach for Pest Disease Prediction International Journal of Information & Computation Technology. ISSN 0974-2239 Volume 4, Number 13 (2014), pp. 1335-1341 International Research Publications House http://www. irphouse.com A Fuzzy Improved

More information

A Brain Computer Interface System For Auto Piloting Wheelchair

A Brain Computer Interface System For Auto Piloting Wheelchair A Brain Computer Interface System For Auto Piloting Wheelchair Reshmi G, N. Kumaravel & M. Sasikala Centre for Medical Electronics, Dept. of Electronics and Communication Engineering, College of Engineering,

More information

Index. Note: Page numbers of article titles are in boldface type.

Index. Note: Page numbers of article titles are in boldface type. Index Note: Page numbers of article titles are in boldface type. A Activity level, in preschoolers, 635 636 ADHD. See Attention-deficit/hyperactivity disorder (ADHD). ADOS. See Autism Diagnostic Observational

More information

Cognitive styles sex the brain, compete neurally, and quantify deficits in autism

Cognitive styles sex the brain, compete neurally, and quantify deficits in autism Cognitive styles sex the brain, compete neurally, and quantify deficits in autism Nigel Goldenfeld 1, Sally Wheelwright 2 and Simon Baron-Cohen 2 1 Department of Applied Mathematics and Theoretical Physics,

More information

Survey of Pathways to Diagnosis and Services

Survey of Pathways to Diagnosis and Services Survey of Pathways to Diagnosis and Services Stephen J. Blumberg, Ph.D. sblumberg@cdc.gov Centers for Disease Control and Prevention National Center for Health Statistics AMCHP Annual Conference February

More information

The Ordinal Nature of Emotions. Georgios N. Yannakakis, Roddy Cowie and Carlos Busso

The Ordinal Nature of Emotions. Georgios N. Yannakakis, Roddy Cowie and Carlos Busso The Ordinal Nature of Emotions Georgios N. Yannakakis, Roddy Cowie and Carlos Busso The story It seems that a rank-based FeelTrace yields higher inter-rater agreement Indeed, FeelTrace should actually

More information

Demystifying DSM 5 Diagnosis

Demystifying DSM 5 Diagnosis 1 New Jersey Center for Tourette Syndrome and Associated Disorders, Inc. Demystifying DSM 5 Diagnosis Colleen Daly Martinez Consultation and Supervision LLC Colleen Martinez, PhD, LCSW, RPT-S March 15,

More information

Autism Diagnosis as a Social Process

Autism Diagnosis as a Social Process Autism Diagnosis as a Social Process An exploration of clinicians diagnostic decision making Supervisors: Dr Ginny Russell Prof Rose McCabe Prof Tamsin Ford Context of study Growing literature on the sociology

More information

Identification of Tissue Independent Cancer Driver Genes

Identification of Tissue Independent Cancer Driver Genes Identification of Tissue Independent Cancer Driver Genes Alexandros Manolakos, Idoia Ochoa, Kartik Venkat Supervisor: Olivier Gevaert Abstract Identification of genomic patterns in tumors is an important

More information

University of the West Indies, Kingston, Jamaica PLEASE SCROLL DOWN FOR ARTICLE

University of the West Indies, Kingston, Jamaica PLEASE SCROLL DOWN FOR ARTICLE This article was downloaded by:[ryerson University] On: 4 June 2008 Access Details: [subscription number 789542258] Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954

More information

A NOVEL VARIABLE SELECTION METHOD BASED ON FREQUENT PATTERN TREE FOR REAL-TIME TRAFFIC ACCIDENT RISK PREDICTION

A NOVEL VARIABLE SELECTION METHOD BASED ON FREQUENT PATTERN TREE FOR REAL-TIME TRAFFIC ACCIDENT RISK PREDICTION OPT-i An International Conference on Engineering and Applied Sciences Optimization M. Papadrakakis, M.G. Karlaftis, N.D. Lagaros (eds.) Kos Island, Greece, 4-6 June 2014 A NOVEL VARIABLE SELECTION METHOD

More information

Rumor Detection on Twitter with Tree-structured Recursive Neural Networks

Rumor Detection on Twitter with Tree-structured Recursive Neural Networks 1 Rumor Detection on Twitter with Tree-structured Recursive Neural Networks Jing Ma 1, Wei Gao 2, Kam-Fai Wong 1,3 1 The Chinese University of Hong Kong 2 Victoria University of Wellington, New Zealand

More information

Inferring Clinical Correlations from EEG Reports with Deep Neural Learning

Inferring Clinical Correlations from EEG Reports with Deep Neural Learning Inferring Clinical Correlations from EEG Reports with Deep Neural Learning Methods for Identification, Classification, and Association using EHR Data S23 Travis R. Goodwin (Presenter) & Sanda M. Harabagiu

More information

UCC EI Underlying Characteristics Checklist Early Intervention 11/19/09. Starting Points. Prevalence of ASD. Starting Points

UCC EI Underlying Characteristics Checklist Early Intervention 11/19/09. Starting Points. Prevalence of ASD. Starting Points Starting Points Underlying Characteristics Checklist EI for ages 0 to 5: Development and Uses Ruth Aspy, Ph.D., Barry G. Grossman, Ph.D., Kathleen Quill, Ed.D., and Nicole Brin, M.A., CCC-SLP Autism spectrum

More information

An Edge-Device for Accurate Seizure Detection in the IoT

An Edge-Device for Accurate Seizure Detection in the IoT An Edge-Device for Accurate Seizure Detection in the IoT M. A. Sayeed 1, S. P. Mohanty 2, E. Kougianos 3, and H. Zaveri 4 University of North Texas, Denton, TX, USA. 1,2,3 Yale University, New Haven, CT,

More information

Universally Beneficial Educational Space Design for Children with Autism; the Research Progression

Universally Beneficial Educational Space Design for Children with Autism; the Research Progression Universally Beneficial Educational Space Design for Children with Autism; the Research Progression Rachna Khare, National Institute of Design, Ahmedabad, India, khare_rachna@hotmail.com Abir Mullick, Georgia

More information

WDHS Curriculum Map Probability and Statistics. What is Statistics and how does it relate to you?

WDHS Curriculum Map Probability and Statistics. What is Statistics and how does it relate to you? WDHS Curriculum Map Probability and Statistics Time Interval/ Unit 1: Introduction to Statistics 1.1-1.3 2 weeks S-IC-1: Understand statistics as a process for making inferences about population parameters

More information