Nicholas Evangelopoulos

Uncovering the intellectual core of the information systems discipline

What is the intellectual core of the information systems discipline? This study uses latent semantic analysis to examine a large body of published IS research in order to address this question. Specifically, the abstracts of all research papers over the time period from 1985 through 2006 published in three top IS research journals-MIS Quarterly, Information Systems Research, and Journal of Management Information Systems-were analyzed. This analysis identified five core research areas: (1) information technology and organizations; (2) IS development; (3) IT and individuals; (4) IT and markets; and (5) IT and groups. Over the time frame of our analysis, these core topics have remained quite stable. However, the specific research themes within each core area have evolved significantly, reflecting research that has focused less on technology development and more on the social context in which information technologies are designed and used. As such, this analysis demonstrates that the information systems academic discipline has maintained a relatively stable research identity that focuses on how IT systems are developed and how individuals, groups, organizations, and markets interact with IT.

Latent Semantic Analysis: Five methodological recommendations

The recent influx in generation, storage, and availability of textual data presents researchers with the challenge of developing suitable methods for their analysis. Latent Semantic Analysis (LSA), a member of a family of methodological approaches that offers an opportunity to address this gap by describing the semantic content in textual data as a set of vectors, was pioneered by researchers in psychology, information retrieval, and bibliometrics. LSA involves a matrix operation called singular value decomposition, an extension of principal component analysis. LSA generates latent semantic dimensions that are either interpreted, if the researchers primary interest lies with the understanding of the thematic structure in the textual data, or used for purposes of clustering, categorization, and predictive modeling, if the interest lies with the conversion of raw text into numerical data, as a precursor to subsequent analysis. This paper reviews five methodological issues that need to be addressed by the researcher who will embark on LSA. We examine the dilemmas, present the choices, and discuss the considerations under which good methodological decisions are made. We illustrate these issues with the help of four small studies, involving the analysis of abstracts for papers published in the European Journal of Information Systems.

Text-mining the voice of the people

Statistical techniques help public leaders turn text in unstructured citizen feedback into responsive e-democracy.

The Use of Latent Semantic Analysis in Operations Management Research

In this article, we introduce the use of Latent Semantic Analysis (LSA) as a technique for uncovering the intellectual structure of a discipline. LSA is an emerging quantitative method for content analysis that combines rigorous statistical techniques and scholarly judgment as it proceeds to extract and decipher key latent factors. We provide a stepwise explanation and illustration for implementing LSA. To demonstrate LSAs ability to uncover the intellectual structure of a discipline, we present a study of the field of Operations Management. We also discuss a number of potential applications of LSA to show how it can be used in empirical Operations Management research, specifically in areas that can benefit from analyzing large volumes of unstructured textual data.

Latent semantic analysis.

This article reviews latent semantic analysis (LSA), a theory of meaning as well as a method for extracting that meaning from passages of text, based on statistical computations over a collection of documents. LSA as a theory of meaning defines a latent semantic space where documents and individual words are represented as vectors. LSA as a computational technique uses linear algebra to extract dimensions that represent that space. This representation enables the computation of similarity among terms and documents, categorization of terms and documents, and summarization of large collections of documents using automated procedures that mimic the way humans perform similar cognitive tasks. We present some technical details, various illustrative examples, and discuss a number of applications from linguistics, psychology, cognitive science, education, information science, and analysis of textual data in general. WIREs Cogn Sci 2013, 4:683-692. doi: 10.1002/wcs.1254 CONFLICT OF INTEREST: The author has declared no conflicts of interest for this article. For further resources related to this article, please visit the WIREs website.

Orthogonal rotations in latent semantic analysis: An empirical study

Abstract The Latent Semantic Analysis (LSA) literature has recently started to address the issue of interpretability of the extracted dimensions. On the software implementation front, recent versions of SAS Text Miner ® started incorporating Varimax rotations. Considering open source software such as R, when it comes to rotation procedures the user has many more options. However, there is a little work in providing guidance for selecting an appropriate rotation procedure. In this paper we further previous research on LSA rotations by introducing two well-known orthogonal rotations, namely Quartimax and Equamax, and comparing them to Varimax. We present a study that empirically tests the influence of the chosen orthogonal rotations on the extraction and interpretation of LSA factors. Our results indicate that, in most cases, Varimax and Equamax produce factors with similar interpretation, while Quartimax tends to produce a single factor. We conclude with recommendations on how these rotation procedures should be used and suggestions for future research. We note that orthogonal rotations can be used to improve the interpretability of other SVD-based models, such as COALS.

The Dual Micro/macro Informing Role of Social Network Sites: Can Twitter Macro Messages Help Predict Stock Prices?

Introduction Emergence and proliferation of social media is one of the most significant trends of the 2000s. Starting with the establishment of Friendster in 2003 as the first large-scale social networking site (Boyd, 2006), a mere 8 years has seen the Internet territory explode with a variety of Social Networking Sites (SNS) which cater to different user categories and are tailored for different purposes. The widespread adoption and popularity of these websites has generated interest from consumers, companies, and researchers alike. Consumers have discovered the value of social network sites and are flocking to them in droves. Social networking sites as a whole almost doubled in unique visitors per month from around 115 million in July of 2007, to just over 200 million in April of 2011 according to one measure (MarketingProfs.com, 2011), and as of June 2012, unique visits to social networking sites have increased by 6% every year (Perez, 2012). Companies are striving to capitalize on the trend of consumer engagement in a variety of ways. Some are using SNS to find new customers (Kulikowski, 2011), to market to existing customers and build their brand (R2Integrated, 2010), and as a new avenue for customer communication for many top U.S. companies (Farfan, 2011). An emerging group of financial traders is closely monitoring SNS for information, crowd pulse, and connection with each other (Harris & Rogers, 2011). The majority of research on SNS is focused on use, adoption, privacy and security issues of SNS (Fogel & Nehmad, 2009; Shen & Khalifa, 2010). SNS generate large volumes of data which in aggregate contain useful information about views and opinions of SNS users. Several studies have investigated the ways to interpret and utilize the information contained in SNS-generated data (e.g., Naaman, Becker, & Gravano, 2011; Taraszow, Aristodemou, Shitta, Laouris, & Arsoy, 2010). This suggests that SNS play a dual role. On the one hand they provide infrastructure for individual informers to deliver messages to individual clients. On the other hand, due to a largely public nature of the individual message exchange, SNS themselves serve as informers providing aggregate information to clients. The goal of this paper is to introduce and validate a framework for understanding the dual role of SNS as platforms for exchanging micro messages and as macro informers. To validate the micro/macro informing framework, we examine if Twitter can serve as a macro informer to the stock market. By applying text mining techniques to SNS messages for the purpose of stock market prediction, this paper represents transdisciplinary research in the fields of information systems and finance, and thus contributes to the goal of Informing Science transdiscipline to reward and encourage research that crosses disciplinary boundaries (Cohen, 2009). In the next section we provide a brief overview of key types of SNS and discuss prior work related to mining the information from SNS sites. We then formally introduce the micro/macro informing framework. We continue with our research methodology and present the results of our analysis. We conclude with a discussion of implications for practice and research. Overview The term social networking site generally refers to those websites and their derived applications that deliver innovative online communication within and among a persons various social networks. SNS typically share a common set of features which include a profile (representation and/or description) for each user, the means to build and manage a personal relational network (i.e., friends, family, acquaintances, etc.), and access to creative methods to communicate with members of their relational network and the online community (Magro, Ryan, Sharp, & Ryan, 2008). SNS have grown in popularity at a tremendous rate in the past several years. Facebook, which has emerged as the dominant SNS, reported 900 million active users in April of 2012 (Goldman, 2012). …

Agenda setting and active audiences in online coverage of human trafficking

ABSTRACT Online news platforms and social media increasingly influence the public agenda on social issues such as human trafficking. Yet despite the popularity of online news and the availability of sophisticated tools for analyzing digital texts, little is known about the relations between news coverage of human trafficking and audiences’ reactions to and interpretations of such coverage. In this paper, we examine journalists’ and commenters’ topic choices in coverage and discussion of human trafficking in the British newspaper The Guardian from 2009 to 2014. We use latent semantic analysis to identify 11 topics discussed by both journalists and readers, and analyze each topic in terms of the degree to which journalists and readers agree or disagree in their topic preferences. We find that four topics were preferred equally by journalists and commenters, four were preferred by journalists, and three were preferred by commenters. Our findings suggest that theories of ‘agenda setting’ and of the ‘active audience’ are not mutually exclusive, and the scope of explanation of each depends partly on the specific topic or subtopic that is analyzed.

Organizational identity, meaning, and values: analysis of social media guideline and policy documents

With the increasing use of social media by students, researchers, administrative staff, and faculty in post-secondary education (PSE), a number of institutions have developed guideline and policy documents to set standards for social media use. In this study we analyze social media guidelines and policies across 250 PSE institutions from 10 countries using latent semantic analysis. This initial finding produced a list of 36 universal topics. Subsequently, chi-squared tests were employed to identify distribution differences of content-related factors between American and Non-American PSE institutions. This analysis offered a high-level summary of unstructured text data on the topic of social media guidance. The results include a comprehensive list of recommendations for developing social media guidelines and policies, and a database of social media guideline and policy documents for the PSE sector and other related organizations.

Determining Process Death Based on Censored Activity Data

This article addresses the problem of estimating the time of apparent death in a binary stochastic process. We show that, when only censored data are available, a fitted logistic regression model may estimate the time of death incorrectly. We improve this estimation by utilizing discrete-event simulation to produce simulated complete time series data. The proposed methodology may be applied to situations where time of death cannot be formally determined and has to be estimated based on prolonged inactivity. As an illustration, we use observed monthly activity patterns from 300 real Open Source Software development projects sampled from Sourceforge.net.