Loading...

Social sciences via network analysis and computation

by Tadej Kanduc (Volume editor)
©2015 Conference proceedings 127 Pages

Summary

In recent years information and communication technologies have gained significant importance in the social sciences. Because there is such rapid growth of knowledge, methods and computer infrastructure, research can now seamlessly connect interdisciplinary fields such as business process management, data processing and mathematics. This study presents some of the latest results, practices and state-of-the-art approaches in network analysis, machine learning, data mining, data clustering and classifications in the contents of social sciences. It also covers various real-life examples such as the analysis of web sentiments, interconnections of methods for text recognition, text analysis and segmentation, information system management and decision support approaches in marketing.

Table Of Contents

  • Cover
  • Title
  • Copyright
  • About the Editor
  • About the Book
  • This eBook can be cited
  • Acknowledgments
  • Table of Contents
  • Preface
  • Text analysis and computations
  • Web Clipping and Sentiment Analysis of Slovenian news articles
  • Machine Learning in Classification of News Portal Articles
  • Network Motifs Analysis of Croatian Literature
  • Toward Network-based Keyword Extraction from Multitopic Web Documents
  • Implementations of methods in practice
  • Automatic invoice capture in small and medium-sized Slovenian enterprises – project report
  • Information System Mirror – Approach How to Analyze Information System Strengths and Weaknesses within Organization
  • Comparison of SAS/STAT Procedures and Variable Neighborhood Search Based Clustering Applied on Telekom Serbia Data
  • Theoretical results on networks and methods
  • Measuring Classification-Tree Comprehensibility
  • On vertex-parity and weak vertex-parity edge-colorings
  • Diameter on some classes of fullerene graphs

Preface

Information and communication technologies (ICT) have gained a significant importance in the field of social sciences in the recent years. Due to rapid growth and development of knowledge, methods and computer infrastructure the research area seamlessly connects interdisciplinary fields such as business process management, data processing and mathematics.

This book includes some of the latest results, practices and novel state-of-the-art approaches in network analysis, machine learning, data mining, data clustering and classifications in the contents of social sciences. It also covers various real-life examples such as analysis of web sentiments, interconnections of methods for text recognition, text analysis and segmentation, information system management and decision support approaches in marketing. The monograph therefore demonstrates how novel approaches using networks analysis and computational methods can contribute to gain new insights and practical applications in social sciences.

In the monograph, twenty-five authors contributed ten scientific papers. The articles are divided into three sections: Text analysis and computations, Implementations of methods in practice and Theoretical results on networks and methods. ← 9 | 10 → ← 10 | 11 →

Jože Bučar, Janez Povh, Andrej Dobrovoljc

Faculty of Information Studies

University of Novo mesto

Sevno 13, 8000 Novo mesto, Slovenia

{joze.bucar, janez.povh, andrej.dobrovoljc}@fis.unm.si

Web Clipping and Sentiment Analysis of Slovenian news articles

Abstract: Web clipping deals with retrieving text and graphic components from web pages. The projects are conducted in collaboration with company Nevtron & Company, d.o.o. that manages the leading IT portal in Slovenia and produces news media, whose role involves experimental testing of proposed solutions. We describe an approach that automates fast and effective collection of data which is crucial for future editorial and business decisions. We have developed solutions that provide a better flexibility to users in selecting, retrieving, extracting and tracking information in publicly available publications on the web. This paper presents retrieval, data pre-processing, sentiment classification, and empirical evaluation of Slovenian news articles enriched with political, business, economic and finance content between September 1st 2007 and December 31st 2013.

Key Words: web clipping, sentiment analysis, sentiment classification, text mining

1.Introduction

An enormous quantity of data is generated on the web daily. We are practically deluged by all kinds of data – scientific, medical, financial, historical, health care, demographic, business, and other. Usually, there are not enough human resources to examine this data. From this chaotic cluster of data we strive to obtain valuable information, which may significantly impact strategic decisions of both business and individuals in the future. The increasing interest in web content has attracted the collaboration among scientists from various fields such as computer science, data mining, machine learning, computational linguistics, graph theory, neural networks, sociology, and psychology. The ability to understand and gain knowledge from text is a key requirement in the goal of artificial intelligence researchers to create machines that simulate the most complex thinking machine in the universe: the human brain.

Relevant information about an organization, its structure, employees, activities, products and services can appear anywhere on the web. “An increasing number of blogs, web pages, newsgroups, forums, chat rooms, etc. has allowed people to express and aggregate their feelings about products, services, events, popularity of ← 11 | 12 → political candidates more intensively.” (Liu 2010, p. 632) Although information on the web can be either true or false, it has significant impact on public opinion and its response. However, more and more business, sale, finance, and other companies are interested in people’s opinion. In a constant battle for success businesses are looking to market their products, identify new opportunities and manage their reputations. The popularity of social media such as social networks, blogs and others has escalated interest in sentiment analysis.

Early text mining activities were concerned primarily with various forms of information retrieval and information summarization, such as indexes, abstracts, and grouping of documents. Later developments in the text mining focused on information extraction. The content and relationship information is extracted from a corpus of documents. “These information extraction operations are performed by tagging each document for the presence of certain content facts or relationships between them.” (Miner 2012, p. 9) “Information extraction consists of an ordered series of steps designed to extract terms, attributes of the terms, facts, and events.” (Sanger and Feldman 2007) Typical text mining tasks include: classification and categorization of texts, sentiment analysis, topic detection, summarization of texts, and a study of relationships between entities in the texts.

In most developed countries automated monitoring of information on the web and other media is an everyday routine, since it improves distinctness, ensures stability and reputation of organizations. There are several successful companies, offering comprehensive solutions from detecting, filtering, classifying, analysing and informing users such as Google Alerts, Web Clipping, etc… In Slovenia automated monitoring of information on the web is not yet widespread.

Details

Pages
127
Year
2015
ISBN (PDF)
9783653058208
ISBN (ePUB)
9783653963588
ISBN (MOBI)
9783653963571
ISBN (Softcover)
9783631665220
DOI
10.3726/978-3-653-05820-8
Language
English
Publication date
2015 (May)
Keywords
computer infrastructure business process management data processing Information and communication technologies
Published
Frankfurt am Main, Berlin, Bern, Bruxelles, New York, Oxford, Wien, 2015. 127 pp., 25 tables, 36 graphs

Biographical notes

Tadej Kanduc (Volume editor)

Tadej Kanduč is a researcher at the Faculty of Information Studies in Novo mesto, Slovenia. His work focuses on discrete and agent based modelling and simulation, optimisation algorithms and parallel programming.

Previous

Title: Social sciences via network analysis and computation
book preview page numper 1
book preview page numper 2
book preview page numper 3
book preview page numper 4
book preview page numper 5
book preview page numper 6
book preview page numper 7
book preview page numper 8
book preview page numper 9
book preview page numper 10
book preview page numper 11
book preview page numper 12
book preview page numper 13
book preview page numper 14
book preview page numper 15
book preview page numper 16
book preview page numper 17
book preview page numper 18
book preview page numper 19
book preview page numper 20
book preview page numper 21
book preview page numper 22
book preview page numper 23
book preview page numper 24
book preview page numper 25
book preview page numper 26
129 pages