Cambria / Hussain Sentic Computing

Techniques, Tools, and Applications
2012
ISBN: 978-94-007-5070-8
Verlag: Springer Netherland
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band 2, 153 Seiten, eBook

Reihe: SpringerBriefs in Cognitive Computation

In this book common sense computing techniques are further developed and applied to bridge the semantic gap between word-level natural language data and the concept-level opinions conveyed by these. In particular, the ensemble application of graph mining and multi-dimensionality reduction techniques is exploited on two common sense knowledge bases to develop a novel intelligent engine for open-domain opinion mining and sentiment analysis. The proposed approach, termed sentic computing, performs a clause-level semantic analysis of text, which allows the inference of both the conceptual and emotional information associated with natural language opinions and, hence, a more efficient passage from (unstructured) textual information to (structured) machine-processable data.

Zielgruppe

Research

Autoren/Hrsg.

Cambria, Erik

Hussain, Amir

Inhaltsverzeichnis

Introduction1.1 Sentic Computing 1.1.1 Motivations 1.1.2 Aims 1.1.3 MethodologyBackground 2.1 Opinion Mining and Sentiment Analysis 2.1.1 The Buzz Mechanism 2.1.2 Origins and Peculiarities 2.1.3 Sub-Tasks2.2 Main Approaches to Opinion Mining 2.2.1 From Heuristics to Discourse Structure 2.2.2 From Coarse to Fine Grained 2.2.3 From Keywords to Concepts2.3 Towards Machines with Common Sense 2.3.1 The Importance of Common Sense 2.3.2 Knowledge Representation 2.3.3 From Logical Inference to Digital Intuition2.4 ConclusionsTechniques3.1 Affective Blending: Enabling Emotion-Sensitive Inference 3.1.1 AffectNet 3.1.2 AffectiveSpace3.2 Affective Categorisation: Modelling Human Emotions 3.2.1 Categorical Versus Dimensional Approaches 3.2.2 The Hourglass of Emotions3.3 Sentic Medoids: Clustering Affective Common Sense Concepts 3.3.1 Partitioning Around Medoids 3.3.2 Centroid Selection3.4 Sentic Activation: A Two-Level Affective Reasoning Framework 3.4.1 Unconscious Reasoning 3.4.2 Conscious Reasoning3.5 Sentic Panalogy: Switching Between Different Ways to Think 3.5.1 Changing Reasoning Strategies 3.5.2 Changing Reasoning Foci3.6 ConclusionsTools4.1 SenticNet: A Semantic Resource for Opinion Mining 4.1.1 Building SenticNet 4.1.2 Working with SenticNet4.2 Sentic Neural Networks: Brain-Inspired Affective Reasoning 4.2.1 Discrete Versus Continuous Approach 4.2.2 Affective Learning4.3 Open Mind Common Sentics: An Emotion-Sensitive IUI 4.3.1 Games for Knowledge Acquisition 4.3.2 Collecting Affective Common Sense Knowledge4.4 Isanette: A Common and Common Sense Knowledge Base 4.4.1 Probase 4.4.2 Building the Instance-Concept Matrix4.5 Opinion Mining Engine: Structuring the Unstructured 4.5.1 Constitutive Modules 4.5.2 Evaluation4.6 ConclusionsApplications5.1 Development of Social Web Systems 5.1.1 Troll Filtering 5.1.2 Social Media Marketing 5.1.3 Sentic Album5.2 Development of HCI Systems 5.2.1 Sentic Avatar 5.2.2 Sentic Chat 5.2.3 Sentic Corner5.3 Development of E-Health Systems 5.3.1 Crowd Validation 5.3.2 Sentic PROMs5.4 ConclusionsConcluding Remarks6.1 Summary of Contributions 6.1.1 Techniques 6.1.2 Tools 6.1.3 Applications6.2 Limitations and Future Work 6.2.1 Limitations 6.2.2 Future Work6.3 ConclusionsReferences