Socio-Affective Computing – serie

This latest volume in the series, Socio-Affective Computing, presents a set of novel approaches to analyze opinionated videos and to extract sentiments and emotions. Textual sentiment analysis framework as discussed in this book contains a novel way of doing sentiment analysis by merging linguistics with machine learning. Fusing textual information with audio and visual cues is found to be extremely useful which improves text, audio and visual based unimodal sentiment analyzer. This volume covers the three main topics of: textual preprocessing and sentiment analysis methods; frameworks to process audio and visual data; and methods of textual, audio and visual features fusion.The inclusion of key visualization and case studies will enable readers to understand better these approaches. Aimed at the Natural Language Processing, Affective Computing and Artificial Intelligence audiences, this comprehensive volume will appeal to a wide readership and will help readers to understand key details on multimodal sentiment analysis.

This book presents a systematic application of recent advances in artificial intelligence (AI) to the problem of asset management. While natural language processing and text mining techniques, such as semantic representation, sentiment analysis, entity extraction, commonsense reasoning, and fact checking have been evolving for decades, finance theories have not yet fully considered and adapted to these ideas.In this unique, readable volume, the authors discuss integrating textual knowledge and market sentiment step-by-step, offering readers new insights into the most popular portfolio optimization theories: the Markowitz model and the Black-Litterman model. The authors also provide valuable visions of how AI technology-based infrastructures could cut the cost of and automate wealth management procedures.This inspiring book is a must-read for researchers and bankers interested in cutting-edge AI applications in finance.

This book presents a systematic application of recent advances in artificial intelligence (AI) to the problem of asset management. While natural language processing and text mining techniques, such as semantic representation, sentiment analysis, entity extraction, commonsense reasoning, and fact checking have been evolving for decades, finance theories have not yet fully considered and adapted to these ideas.In this unique, readable volume, the authors discuss integrating textual knowledge and market sentiment step-by-step, offering readers new insights into the most popular portfolio optimization theories: the Markowitz model and the Black-Litterman model. The authors also provide valuable visions of how AI technology-based infrastructures could cut the cost of and automate wealth management procedures.This inspiring book is a must-read for researchers and bankers interested in cutting-edge AI applications in finance.

This book presents a synthetic analysis about the characteristics of time expressions and named entities, and some proposed methods for leveraging these characteristics to recognize time expressions and named entities from unstructured text. For modeling these two kinds of entities, the authors propose a rule-based method that introduces an abstracted layer between the specific words and the rules, and two learning-based methods that define a new type of tagging scheme based on the constituents of the entities, different from conventional position-based tagging schemes that cause the problem of inconsistent tag assignment. The authors also find that the length-frequency of entities follows a family of power-law distributions. This finding opens a door, complementary to the rank-frequency of words, to understand our communicative system in terms of language use.

This book presents a synthetic analysis about the characteristics of time expressions and named entities, and some proposed methods for leveraging these characteristics to recognize time expressions and named entities from unstructured text. For modeling these two kinds of entities, the authors propose a rule-based method that introduces an abstracted layer between the specific words and the rules, and two learning-based methods that define a new type of tagging scheme based on the constituents of the entities, different from conventional position-based tagging schemes that cause the problem of inconsistent tag assignment. The authors also find that the length-frequency of entities follows a family of power-law distributions. This finding opens a door, complementary to the rank-frequency of words, to understand our communicative system in terms of language use.

In this book, the authors systematically and comprehensively discuss textual emotion classification by using deep broad learning. Since broad learning possesses certain advantages such as simple network structure, short training time and strong generalization ability, it is a new and promising framework for textual emotion classification in artificial intelligence. As a result, how to combine deep and broad learning has become a new trend of textual emotion classification, a booming topic in both academia and industry.For a better understanding, both quantitative and qualitative results are present in figures, tables, or other suitable formats to give the readers the broad picture of this topic along with unique insights of common sense and technical details, and to pave a solid ground for their forthcoming research or industry applications. In a progressive manner, the readers will gain exclusive knowledge in textual emotion classification using deep broad learning and be inspired to further investigate this underexplored domain.With no other similar book existing in the literature, the authors aim to make the book self-contained for newcomers, only a few prerequisites being expected from the readers. The book is meant as a reference for senior undergraduates, postgraduates, scientists and researchers interested to have a quick idea of the foundations and research progress of security and privacy in federated learning, and it can equally well be used as a textbook by lecturers, tutors, and undergraduates.

In this book, the authors systematically and comprehensively discuss textual emotion classification by using deep broad learning. Since broad learning possesses certain advantages such as simple network structure, short training time and strong generalization ability, it is a new and promising framework for textual emotion classification in artificial intelligence. As a result, how to combine deep and broad learning has become a new trend of textual emotion classification, a booming topic in both academia and industry.For a better understanding, both quantitative and qualitative results are present in figures, tables, or other suitable formats to give the readers the broad picture of this topic along with unique insights of common sense and technical details, and to pave a solid ground for their forthcoming research or industry applications. In a progressive manner, the readers will gain exclusive knowledge in textual emotion classification using deep broad learning and be inspired to further investigate this underexplored domain.With no other similar book existing in the literature, the authors aim to make the book self-contained for newcomers, only a few prerequisites being expected from the readers. The book is meant as a reference for senior undergraduates, postgraduates, scientists and researchers interested to have a quick idea of the foundations and research progress of security and privacy in federated learning, and it can equally well be used as a textbook by lecturers, tutors, and undergraduates.

This volume presents a knowledge-based approach to concept-level sentiment analysis at the crossroads between affective computing, information extraction, and common-sense computing, which exploits both computer and social sciences to better interpret and process information on the Web. Concept-level sentiment analysis goes beyond a mere word-level analysis of text in order to enable a more efficient passage from (unstructured) textual information to (structured) machine-processable data, in potentially any domain. Readers will discover the following key novelties, that make this approach so unique and avant-garde, being reviewed and discussed:•    Sentic Computing's multi-disciplinary approach to sentiment analysis-evidenced by the concomitant use of AI, linguistics and psychology for knowledge representation and inference•    Sentic Computing’s shift from syntax to semantics-enabled by the adoption of the bag-of-concepts model instead of simply counting word co-occurrence frequencies in text•    Sentic Computing's shift from statistics to linguistics-implemented by allowing sentiments to flow from concept to concept based on the dependency relation between clausesThis volume is the first in the Series Socio-Affective Computing edited by Dr Amir Hussain and Dr Erik Cambria and will be of interest to researchers in the fields of socially intelligent, affective and multimodal human-machine interaction andsystems.

The objective of this monograph is to improve the performance of the sentiment analysis model by incorporating the semantic, syntactic and common-sense knowledge. This book proposes a novel semantic concept extraction approach that uses dependency relations between words to extract the features from the text. Proposed approach combines the semantic and common-sense knowledge for the better understanding of the text. In addition, the book aims to extract prominent features from the unstructured text by eliminating the noisy, irrelevant and redundant features. Readers will also discover a proposed method for efficient dimensionality reduction to alleviate the data sparseness problem being faced by machine learning model. Authors pay attention to the four main findings of the book :-Performance of the sentiment analysis can be improved by reducing the redundancy among the features. Experimental results show that minimum Redundancy Maximum Relevance (mRMR) feature selection technique improves the performance of the sentiment analysis by eliminating the redundant features.- Boolean Multinomial Naive Bayes (BMNB) machine learning algorithm with mRMR feature selection technique performs better than Support Vector Machine (SVM) classifier for sentiment analysis.- The problem of data sparseness is alleviated by semantic clustering of features, which in turn improves the performance of the sentiment analysis.- Semantic relations among the words in thetext have useful cues for sentiment analysis. Common-sense knowledge in form of ConceptNet ontology acquires knowledge, which provides a better understanding of the text that improves the performance of the sentiment analysis.

The idea of this book is to establish a new scientific discipline, “noology,” under which a set of fundamental principles are proposed for the characterization of both naturally occurring and artificial intelligent systems. The methodology adopted in Principles of Noology for the characterization of intelligent systems, or “noological systems,” is a computational one, much like that of AI. Many AI devices such as predicate logic representations, search mechanisms, heuristics, and computational learning mechanisms are employed but they are recast in a totally new framework for the characterization of noological systems. The computational approach in this book provides a quantitative and high resolution understanding of noological processes, and at the same time the principles and methodologies formulated are directly implementable in AI systems.In contrast to traditional AI that ignores motivational and affective processes, under the paradigm of noology,motivational and affective processes are central to the functioning of noological systems and their roles in noological processes are elucidated in detailed computational terms. In addition, a number of novel representational and learning mechanisms are proposed, and ample examples and computer simulations are provided to show their applications. These include rapid effective causal learning (a novel learning mechanism that allows an AI/noological system to learn causality with a small number of training instances), learning of scripts that enables knowledge chunking and rapid problem solving, and learning of heuristics that further accelerates problem solving. Semantic grounding allows an AI/noological system to “truly understand” the meaning of the knowledge it encodes. This issue is extensively explored.This is a highly informative book providing novel and deep insights into intelligent systems which is particularly relevant to both researchers and students of AI and the cognitive sciences.

The core message of this book is: computer games best realise affective interaction. This book brings together contributions from specialists in affective computing, game studies, game artificial intelligence, user experience research, sensor technology, multi-modal interfaces and psychology that will advance the state-of-the-art in player experience research; affect modelling, induction, and sensing; affect-driven game adaptation and game-based learning and assessment. In 3 parts the books covers Theory, Emotion Modelling and Affect-Driven Adaptation, and Applications. This book will be of interest to researchers and scholars in the fields of game research, affective computing, human computer interaction, and artificial intelligence.

Sentiment analysis research has been started long back and recently it is one of the demanding research topics.  Research activities on Sentiment Analysis in natural language texts and other media are gaining ground with full swing. But, till date, no concise set of factors has been yet defined that really affects how writers’ sentiment i.e., broadly human sentiment is expressed, perceived, recognized, processed, and interpreted in natural languages. The existing reported solutions or the available systems are still far from perfect or fail to meet the satisfaction level of the end users. The reasons may be that there are dozens of conceptual rules that govern sentiment and even there are possibly unlimited clues that can convey these concepts from realization to practical implementation. Therefore, the main aim of this book is to provide a feasible research platform to our ambitious researchers towards developing the practical solutions that will be indeed beneficial for our society,business and future researches as well.

This book presents a summary of the multimodal analysis of user-generated multimedia content (UGC). Several multimedia systems and their proposed frameworks are also discussed. First, improved tag recommendation and ranking systems for social media photos, leveraging both content and contextual information, are presented. Next, we discuss the challenges in determining semantics and sentics information from UGC to obtain multimedia summaries. Subsequently, we present a personalized music video generation system for outdoor user-generated videos. Finally, we discuss approaches for multimodal lecture video segmentation techniques. This book also explores the extension of these multimedia system with the use of heterogeneous continuous streams.

The abundance of text available in social media and health-related forums and blogs have recently attracted the interest of the public health community to use these sources for opinion mining. This book presents a lexicon-based approach to sentiment analysis in the bio-medical domain, i.e., WordNet for Medical Events (WME). This book gives an insight in handling unstructured textual data and converting it to structured machine-processable data in the bio-medical domain.The readers will discover the following key novelties:1) development of a bio-medical lexicon: WME expansion and WME enrichment with additional features.;2) ensemble of machine learning and computational creativity;3) development of microtext analysis techniques to overcome the inconsistency in social communication.It will be of interest to researchers in the fields of socially-intelligent human-machine interaction and biomedical text mining

This volume presents a knowledge-based approach to concept-level sentiment analysis at the crossroads between affective computing, information extraction, and common-sense computing, which exploits both computer and social sciences to better interpret and process information on the Web. Concept-level sentiment analysis goes beyond a mere word-level analysis of text in order to enable a more efficient passage from (unstructured) textual information to (structured) machine-processable data, in potentially any domain. Readers will discover the following key novelties, that make this approach so unique and avant-garde, being reviewed and discussed:•    Sentic Computing's multi-disciplinary approach to sentiment analysis-evidenced by the concomitant use of AI, linguistics and psychology for knowledge representation and inference•    Sentic Computing’s shift from syntax to semantics-enabled by the adoption of the bag-of-concepts model instead of simply counting word co-occurrence frequencies in text•    Sentic Computing's shift from statistics to linguistics-implemented by allowing sentiments to flow from concept to concept based on the dependency relation between clausesThis volume is the first in the Series Socio-Affective Computing edited by Dr Amir Hussain and Dr Erik Cambria and will be of interest to researchers in the fields of socially intelligent, affective and multimodal human-machine interaction andsystems.

The objective of this monograph is to improve the performance of the sentiment analysis model by incorporating the semantic, syntactic and common-sense knowledge. This book proposes a novel semantic concept extraction approach that uses dependency relations between words to extract the features from the text. Proposed approach combines the semantic and common-sense knowledge for the better understanding of the text. In addition, the book aims to extract prominent features from the unstructured text by eliminating the noisy, irrelevant and redundant features. Readers will also discover a proposed method for efficient dimensionality reduction to alleviate the data sparseness problem being faced by machine learning model. Authors pay attention to the four main findings of the book :-Performance of the sentiment analysis can be improved by reducing the redundancy among the features. Experimental results show that minimum Redundancy Maximum Relevance (mRMR) feature selection technique improves the performance of the sentiment analysis by eliminating the redundant features.- Boolean Multinomial Naive Bayes (BMNB) machine learning algorithm with mRMR feature selection technique performs better than Support Vector Machine (SVM) classifier for sentiment analysis.- The problem of data sparseness is alleviated by semantic clustering of features, which in turn improves the performance of the sentiment analysis.- Semantic relations among the words in thetext have useful cues for sentiment analysis. Common-sense knowledge in form of ConceptNet ontology acquires knowledge, which provides a better understanding of the text that improves the performance of the sentiment analysis.

The core message of this book is: computer games best realise affective interaction. This book brings together contributions from specialists in affective computing, game studies, game artificial intelligence, user experience research, sensor technology, multi-modal interfaces and psychology that will advance the state-of-the-art in player experience research; affect modelling, induction, and sensing; affect-driven game adaptation and game-based learning and assessment. In 3 parts the books covers Theory, Emotion Modelling and Affect-Driven Adaptation, and Applications. This book will be of interest to researchers and scholars in the fields of game research, affective computing, human computer interaction, and artificial intelligence.

Sentiment analysis research has been started long back and recently it is one of the demanding research topics.  Research activities on Sentiment Analysis in natural language texts and other media are gaining ground with full swing. But, till date, no concise set of factors has been yet defined that really affects how writers’ sentiment i.e., broadly human sentiment is expressed, perceived, recognized, processed, and interpreted in natural languages. The existing reported solutions or the available systems are still far from perfect or fail to meet the satisfaction level of the end users. The reasons may be that there are dozens of conceptual rules that govern sentiment and even there are possibly unlimited clues that can convey these concepts from realization to practical implementation. Therefore, the main aim of this book is to provide a feasible research platform to our ambitious researchers towards developing the practical solutions that will be indeed beneficial for our society,business and future researches as well.

This book presents a summary of the multimodal analysis of user-generated multimedia content (UGC). Several multimedia systems and their proposed frameworks are also discussed. First, improved tag recommendation and ranking systems for social media photos, leveraging both content and contextual information, are presented. Next, we discuss the challenges in determining semantics and sentics information from UGC to obtain multimedia summaries. Subsequently, we present a personalized music video generation system for outdoor user-generated videos. Finally, we discuss approaches for multimodal lecture video segmentation techniques. This book also explores the extension of these multimedia system with the use of heterogeneous continuous streams.

The abundance of text available in social media and health-related forums and blogs have recently attracted the interest of the public health community to use these sources for opinion mining. This book presents a lexicon-based approach to sentiment analysis in the bio-medical domain, i.e., WordNet for Medical Events (WME). This book gives an insight in handling unstructured textual data and converting it to structured machine-processable data in the bio-medical domain.The readers will discover the following key novelties:1) development of a bio-medical lexicon: WME expansion and WME enrichment with additional features.;2) ensemble of machine learning and computational creativity;3) development of microtext analysis techniques to overcome the inconsistency in social communication.It will be of interest to researchers in the fields of socially-intelligent human-machine interaction and biomedical text mining