On the platform but will they buy? Predicting customers' purchase behavior using deep learning

2021 | Decision Support Systems | Citations: 7

Authors: Chaudhuri, Neha; Gupta, Gaurav; Vamsi, Vallurupalli; Bose, Indranil

Abstract: A thorough understanding of online customer’s purchase behavior will directly bo ... Expand

Abstract: A thorough understanding of online customer’s purchase behavior will directly boost e-commerce business performance. Existing studies have overtly focused on purchase intention and used sales rank as a natural proxy, which however has limited business application. Additionally, intention to purchase does not necessarily convert to actual retail purchases. We aim to further our understanding of online customer’s purchase behavior for an ecommerce platform by predicting the same using deep learning techniques, on a large multidimensional data sample of more than 50,000 unique web sessions. This study used two distinct sets of variables, i.e., platform engagement and customer characteristics, as key predictors of online purchases by retail customers. We further compared the predictive capability of our deep learning method with other widely used machine learning techniques for prediction, including Decision Tree, Random Forest, Support Vector Machines, and Artificial Neural Networks. We found that the deep learning technique outperformed the machine learning techniques when applied to the same dataset. These analyses will help platform designers plan for more platform engage­ ments while simultaneously expanding the academic understanding of purchase prediction for online e-com­ merce platforms. Collapse

Topics: electronic commerce missing data open source Google+ lean manufacturing

Methods: machine learning deep learning sensitivity analysis descriptive statistic support vector machine

Synthesizing and Integrating Research on IT-Based Value Cocreation: A Meta-Analysis

2020 | Journal of the Association for Information Systems | Citations: 0

Authors: Mandrella, Markus; Trang, Simon; Kolbe, Lutz

Abstract: IT value research has witnessed growing interest in the use of joint IT resource ... Expand

Abstract: IT value research has witnessed growing interest in the use of joint IT resources and capabilities following recent shifts in market competition from the firm to the network level. Despite research efforts in this domain, there remain substantial inconsistencies in the IT value cocreation literature regarding the effect of interorganizational IT on business value and the role of methodological and contextual factors. Drawing on the resource-based view and the relational view of the firm, we conducted a meta-analysis to synthesize and integrate the body of knowledge of IT-based value cocreation. Our analysis of 80 studies, encompassing 21,843 observations, highlights the value-generating effect of four interorganizational IT capabilities: IT-based relation-specific assets, IT-based knowledge sharing, IT-based complementary capabilities, and IT-based governance. Insights from our preliminary meta-analysis reveal that contradictory findings are driven by the conceptualization of IT variables as interorganizational IT resources. A further moderator meta-analysis explains divergent empirical findings in the literature. We find that the use of relational-level value and perceptual measures, use of single respondents, and the context of developing countries and supply chain and networked interdependencies result in larger estimates of business value. In contrast, the use of network-level, firm-level, and objective measures; use of matched-pair approaches; and the context of developed countries and pooled interdependencies result in smaller estimates. Overall, this paper provides clarity and structure to the current understanding of the research field by providing explanations for inconsistent findings as well as a foundation for future research and theory development. Collapse

Topics: organizational value interorganizational system information technology capability participatory design IS value

Methods: design methodology meta analysis theory development descriptive statistic mediation analysis

Theories: resource based view of the firm

Examining Real Options Exercise Decisions in Information Technology Investments

2017 | Journal of the Association for Information Systems | Citations: 0

Authors: Khan, Sarah; Zhao, Kexin; Kumar, Ram; Stylianou, Antonis

Abstract: Researchers have advocated real options thinking (ROT) for evaluating and managi ... Expand

Abstract: Researchers have advocated real options thinking (ROT) for evaluating and managing risky IT investments to account for managerial decision flexibility. Effective ROT is a three-step process that requires managers to recognize, value, and exercise options embedded in IT projects. Prior research has illustrated the existence of managerial bias in the recognizing and valuing real options. However, little research has examined real options exercise decisions. Hence, we use prospect theory to examine whether IT managers demonstrate systematic biases while exercising real options in IT projects and portfolios. We also study whether one can control or mitigate such biases. We found evidence of biased (suboptimal) real option exercise decisions in IT projects and in IT portfolios. However, we found differences in biased decision making between a single project and a portfolio scenario. We also found that project scale and real option type influenced vulnerability of a project to biased decision making. In addition, simplifying the presentation of the net effects of real options exercise decisions can help reduce bias, especially for large project portfolios. We discuss the implications of these results on theory and practice. Collapse

Topics: real option IT manager IT project project management decision making

Methods: survey repeated measures ANOVA parametric test statistical hypothesis test Cochran's Q test

Theories: prospect theory loss aversion behavioral economics economic theory real options theory

An empirical evaluation of the comprehensibility of decision table, tree and rule based predictive models

2011 | Decision Support Systems | Citations: 25

Authors: Huysmans, Johan; Dejaeger, Karel; Mues, Christophe; Vanthienen, Jan; Baesens, Bart

Abstract: An important objective of data mining is the development of predictive models. B ... Expand

Abstract: An important objective of data mining is the development of predictive models. Based on a number of observations, a model is constructed that allows the analysts to provide classifications or predictions for new observations. Currently, most research focuses on improving the accuracy or precision of these models and comparatively little research has been undertaken to increase their comprehensibility to the analyst or end-user. This is mainly due to the subjective nature of ‘comprehensibility’, which depends on many factors outside the model, such as the user's experience and his/her prior knowledge. Despite this influence of the observer, some representation formats are generally considered to be more easily interpretable than others. In this paper, an empirical study is presented which investigates the suitability of a number of alternative representation formats for classification when interpretability is a key requirement. The formats under consideration are decision tables, (binary) decision trees, propositional rules, and oblique rules. An end-user experiment was designed to test the accuracy, response time, and answer confidence for a set of problem-solving tasks involving the former representations. Analysis of the results reveals that decision tables perform significantly better on all three criteria, while post-test voting also reveals a clear preference of users for decision tables in terms of ease of use. Collapse

Topics: ease of use task productivity problem solving

Methods: experiment decision tree classification nonparametric test Bonferroni correction parametric test

Theories: cognitive fit theory

Commercial Internet filters: Perils and opportunities

2010 | Decision Support Systems | Citations: 2

Authors: Chou, Chen-Huei; Sinha, Atish P.; Zhao, Huimin

Abstract: Organizations are becoming increasingly aware of Internet abuse in the workplace ... Expand

Abstract: Organizations are becoming increasingly aware of Internet abuse in the workplace. Such abuse results in loss of workers' productivity, network congestion, security risks, and legal liabilities. To address this problem, organizations have started to adopt Internet usage policies, management training, and filtering software. Several commercial Internet filters are experiencing an increasing number of organizational adoptions. These products mainly rely on black lists, white lists, and keyword/profile matching to filter out undesired web pages. In this paper, we describe three top-ranked commercial Internet filters – CYBERSitter, Net Nanny, and CyberPatrol – and evaluate their performance in the context of an Internet abuse problem. We then propose a text mining approach to address the problem and evaluate its performance using six different classification algorithms: naïve Bayes, multinominal naïve Bayes, support vector machine, decision tree, k-nearest neighbor, and neural network. The evaluation results point to the perils of using commercial Internet filters on one hand, and to the prospects of using text mining on the other. The proposed text mining approach outperforms the commercial filters. We discuss the possible reasons for the relatively poor performance of the filters and the steps that could be taken to improve their performance. Collapse

Topics: web page IT security threat internet filter internet technology software developer

Methods: natural language processing decision tree classification naïve bayes computational algorithm classification method

The Last Mile in an Electronic Commerce Business Model - Service Expectations of Austrian Online Shoppers

2005 | European Conference On Information Systems | Citations: 0

Authors: Madlberger, Maria; Sester, Antje

Abstract: Many existing research approaches on business models in e-commerce are on a high ... Expand

Abstract: Many existing research approaches on business models in e-commerce are on a high and aggregated level. Furthermore, there are few empirical insights into the appropriateness of chosen e-commerce business models. The paper at hand focuses on the empirical analysis of the last mile issue in b2c ecommerce. For this purpose a research framework for e-commerce business models is presented that serves as a basis for empirical analyses as well as for managerial decisions. The framework takes strategic decisions and conditions of an e-commerce business model into account. Based on this framework the specific issue of last mile logistics is empirically analyzed. Relevant action alternatives for physical distribution are investigated according to their appropriateness from consumers’ point of view. The empirical study has been carried out on behalf of 180 face-to-face interviews. Results indicate that the product category has a strong impact on customers’ expectations on an electronic retailer’s last mile logistics service whereas the customers’ flexibility in terms of time and location obviously does not influence their requirements for logistical services. Collapse

Topics: logistics management electronic commerce willingness to pay business model business-to-consumer

Methods: parametric test survey ANOVA personal interview Cochran's Q test

The Repertory Grid Technique: A Method for the Study of Cognition in Information Systems

2002 | Management Information Systems Quarterly | Citations: 168

Authors: Tan, Felix B.; Hunter, M. Gordon

Abstract: Recent studies have confirmed the importance of understanding the cognition of u ... Expand

Abstract: Recent studies have confirmed the importance of understanding the cognition of users and information systems (IS) professionals. These works agree that organizational cognition is far too critical to be ignored as it can impact on IS outcomes. While cognition has been considered in a variety of IS contexts, no specific methodology has dominated. A theory and method suitable to the study of cognition defined as personal constructs that individuals use to understand IT in organizations--is Kelly's (1955) personal construct theory and its cognitive mapping tool known as the repertory grid (RepGrid). This article expounds on the potential of this technique to IS researchers by considering the variety of ways the RepGrid may be employed. The flexibility of the RepGrid is illustrated by examining published studies in IS. The diagnostic qualities of the RepGrid and its mapping outcomes can be used for practical intervention at the individual and organizational levels. Collapse

Topics: IT workforce organizational context IT development project IT project

Methods: qualitative interview ANOVA cluster analysis Cochran's Q test parametric test

Theories: personal construct theory