Recommender Systems for Medicine and Music (Studies in Computational Intelligence, 946)

Description

Music recommendation systems are becoming more and more popular. The increasing amount of personal data left by users on social media contributes to more accurate inference of the user’s musical preferences and the same to quality of personalized systems. Health recommendation systems have become indispensable tools in decision making processes in the healthcare sector. Their main objective is to ensure the availability of valuable information at the right time by ensuring information quality, trustworthiness, authentication, and privacy concerns. Medical doctors deal with various kinds of diseases in which the music therapy helps to improve symptoms. Listening to music may improve heart rate, respiratory rate, and blood pressure in people with heart disease. Sound healing therapy uses aspects of music to improve physical and emotional health and well-being. The book presents a variety of approaches useful to create recommendation systems in healthcare, music, and in music therapy. Preface Acknowledgements Contents Contributors Recommendation Systems in Healthcare 1 Introduction to Recommender Systems 2 Recommender System Classification 2.1 Collaborative Filtering 2.2 Content-Based Recommender Systems 2.3 Knowledge-Based Recommender Systems 2.4 Group Recommender Systems 2.5 Hybrid Recommender Systems References Recommender Systems in Healthcare: A Socio-Technical Systems Approach 1 Introduction 2 Recommender Systems 2.1 Technological and Social Issues in Recommender Systems 2.2 Applications of Recommender Systems in Healthcare 3 Socio-Technical Systems and Their Design 3.1 Healthcare System as a Socio-Technical System 3.2 Autonomous Agents in a Socio-Technical System 3.3 Workload Analysis in a Socio-Technical System 4 Design Considerations for RS in Healthcare: A Socio-Technical Approach 4.1 Technical Agents: Regulations in Healthcare 4.2 Human Agents: Empowerment, Patient- and Relationship-Centered Care 4.3 Human Agents: Human-Driven Care 5 Discussion References Computer Methods for Localization of the Subthalamic Nucleus During Deep Brain Stimulation Surgeries for Treatment of Parkinson Disease 1 Description of the Problem 2 Characteristics of Microelectrode Recorded Signal 2.1 Action Potentials 2.2 Background Activity 3 Attributes 3.1 Spike Activity-Based Attributes 3.2 Preprocessing for Background Activity-Based Attributes 3.3 Background Activity-Based Attributes 3.4 Moving Average-Based Attributes 3.5 Temporal Attributes 3.6 Narrow Frequency Band-Based Attributes 4 Evaluation and Interpretation 4.1 Spike Activity-Based Attributes 4.2 Narrow Frequency Band-Based Attributes 4.3 Attributes Based-On and Derived-From Background Activity 5 Conclusions References Stutter Detection and Remediation in Speech 1 Introduction and Background 2 Stutter Detection System 2.1 Identifying Potentially-Undesired Audio Segments 2.2 Labeling Potentially Undesired Segments 2.3 Constructing the Classifiers 3 Experiments and Results 4 Crowdsourcing Stutter Labels Web-Platform 4.1 System Overview 4.2 Creating a Crowdsourcing Study 4.3 Logic Behind Segment-Worker Label Assignment 5 Conclusion References Personalizing Patients to Enable Shared Decision Making 1 Introduction 2 Healthcare Cost and Utilization Project (H-CUP) 3 Predicting Medical Outcomes 4 Treatment Plan Visualization 5 Personalizing Treatment Plan 6 Conclusion References An LSTM-based Approach for Insulin and Carbohydrate Recommendations in Type 1 Diabetes Self-Management 1 Introduction and Motivation 2 Three Recommendation Scenarios 3 Baseline Models and Neural Architectures 4 Using the OhioT1DM Dataset for Recommendation Examples 4.1 From Meals and Bolus Events to Recommendation Examples 5 Experimental Methodology and Results 5.1 Experimental Results 6 Conclusion References Music Recommendation Systems: A Survey 1 Introduction 2 Music Recommendation Systems 3 Personalization of Music Recommendation Systems 3.1 Emotions 3.2 Personality 3.3 Social Context 3.4 New Interfaces 3.5 Automatic Music Generation 4 Summary References Repeated Listens in the Music Discovery Process 1 The Long-Tail Problem 2 Causes of the Long-Tail Problem 3 Potential Solutions to the Long-Tail Problem 4 Effect of Repeated Listens on Memory and Liking 5 Music Familiarity and Its Influence on Listening Behavior 6 Aspects of Familiarity 7 Familiarity and Emotions 8 How Familiarity Influences Listening Behavior 9 Music Listening Study 10 Experiment Methodology 11 Results and Discussion 12 Conclusions and Future Work References What Songs We Listen to Together: Automatic Music Selection for Groups 1 Introduction 2 Issues in Group Music Selection 3 Existing Works 3.1 MusicFX ch9McCarthyCSCW98 3.2 Flytrap ch9CrossenIUI2002 3.3 GroupFun ch9PopescuCHI2012 3.4 BlueMusic ch9MahatoCHI2008 3.5 Other Works 4 Case Study: A Bluetooth-Based Music Selection Application 4.1 Concept 4.2 System Design 4.3 Implementation 4.4 Evaluation of Preferences for Owned Songs 4.5 Evaluation of Preferences for Unowned Songs 5 Discussion 6 Conclusion References Body Data for Music Information Retrieval Tasks 1 Music and the Body 2 Music Content Retrieval 2.1 Audio Content-Based Music Retrieval 2.2 Content Retrieval Using Melody 3 Body Data 3.1 Body Data and Music Creation 3.2 Body Data for Music Retrieval 4 Search and Retrieval Algorithms 4.1 Multimodal Retrieval 4.2 EEG Data for Music Information Retrieval 4.3 Motion Capture 5 Methods for Multimodal Retrieval 5.1 Time Series Analysis 5.2 Canonical Correlation Analysis 6 Structures of Multimodal Retrieval 6.1 Granularity and Specificity 6.2 Intermediate Domain Representation 7 Conclusions References Music Recommendation Based on Emotion Tracking of Musical Performances 1 Introduction 2 Related Work 3 Preparatory Activities for Finding Similarities 3.1 Annotation of Music Data for Regressor Training 3.2 Training Regressors for Emotion Prediction 3.3 Automatic Alignment of Different Performances 4 Results of Emotion Tracking 4.1 Arousal and Valence Trajectories 4.2 Visual Emotion Tracking on the A-V Plane 4.3 Visual Emotion Tracking on the A-V Plane After Ranking 5 Similarity Findings 5.1 Metrics of Similarity 5.2 Joining Arousal and Valence Sequences 5.3 Results of Calculating Similarity Matrices 6 Evaluation 6.1 Ground Truth Similarity Matrix 6.2 Evaluation Results 7 Conclusions References Music and Healthcare Recommendation Systems 1 Introduction 2 Healthcare Recommendation Systems 3 Systems for Recommending Music in Medicine 4 The Influence of Music on Health 5 Summary References Emotion-Based Music Recommender System for Tinnitus Patients (EMOTIN) 1 Introduction 1.1 Tinnitus 1.2 Tinnitus Retraining Therapy 1.3 Music Therapy 2 Background 2.1 RECTIN—Recommender for Tinnitus 2.2 eTRT—Electronic Tinnitus Retraining Therapy 3 Methods 3.1 Music Recommendations 3.2 Music Therapy Protocol 3.3 Detecting Affective States 3.4 Emotion Model 3.5 Music Emotion Recognition 3.6 Audio Notching 3.7 Emotion-Based Music Recommender Model 3.8 Recommendation Algorithm 4 Results 4.1 Audio Feature Selection for Music Emotion Recognition 4.2 Regression Models for Music Emotion Recognition 5 Discussion References A Model of Typhlo Music Therapy in Educational and Rehabilitation Work with Visually Impaired Persons 1 Introduction 2 Music—The Most Accessible of All the Fine Arts 3 The Concept of Typhlo Music Therapy 4 Typhlo Music Therapeutic Procedure 5 Acoustic Material 6 Music Therapy Sessions 7 Conclusions 8 Notes on the Authors References