Getting Personal: A Deep Learning Artifact for Text-Based Measurement of Personality

2023 | Information Systems Research | Citations: 0

Authors: Yang, Kai; Lau, Raymond Y. K.; Abbasi, Ahmed

Abstract: Analysts, managers, and policymakers are interested in predictive analytics capa ... Expand

Abstract: Analysts, managers, and policymakers are interested in predictive analytics capable of offering better foresight. It is generally accepted that in forecasting scenarios involving organizational policies or consumer decision making, personal characteristics, including personality, may be an important predictor of downstream outcomes. The inclusion of personality features in forecasting models has been hindered by the fact that traditional measurement mechanisms are often infeasible. Text-based personality detection has garnered attention because of the public availability of digital textual traces. However, the text machine learning space has bifurcated into two branches: feature-based methods relying on manually crafted human intuition, or deep learning language models that leverage big data and compute, the main commonality being that neither branch generates accurate personality assessments, thereby making personality measures infeasible for downstream forecasting applications. In this study, we propose DeepPerson, a design artifact for text-based personality detection that bridges these two branches by leveraging concepts from relevant psycholinguistic theories in conjunction with advanced deep learning strategies. DeepPerson incorporates novel transfer learning and hierarchical attention network methods that use psychological concepts and data augmentation in conjunction with person-level linguistic information. We evaluate the utility of the proposed artifact using an extensive design evaluation on three personality data sets in comparison with state-of-the-art methods proposed in academia and industry. DeepPerson can improve detection of personality dimensions by 10–20 percentage points relative to the best comparison methods. Using case studies in the finance and health domains, we show that more accurate text-based personality detection can translate into significant improvements in downstream applications such as forecasting future firm performance or predicting pandemic infection rates. Our findings have important implications for research at the intersection of design and data science, and practical implications for managers focused on enabling, producing, or consuming predictive analytics.History: Olivia Sheng, Senior Editor; Huimin Zhao, Associate Editor.Funding: This work was supported by Oracle for Research (NLP for the Greater Good) and the U.S. National Science Foundation’s Division of Information and Intelligent Systems [Grants BDS-1636933 and IIS-1816504]. The work was also partly supported by the Research Grants Council of the Hong Kong Special Administrative Region [Project: CityU 11507219] and the City University of Hong Kong SRG [Project: 7005780].Supplemental Material: The online appendices are available at https://doi.org/10.1287/isre.2022.1111. Collapse

Topics: personality Twitter social media pandemic Google+

Methods: machine learning BERT design science deep learning computational algorithm

Theories: big five model

Spoiled for Choice? Personalized Recommendation for Healthcare Decisions: A Multiarmed Bandit Approach

2023 | Information Systems Research | Citations: 3

Authors: Zhou, Tongxin; Wang, Yingfei; Yan, Lu (Lucy); Tan, Yong

Abstract: Online healthcare platforms provide users with various intervention programs to ... Expand

Abstract: Online healthcare platforms provide users with various intervention programs to promote personal wellness. Given the many options available, it’s often difficult for individuals to decide in which intervention to participate, especially when they lack the experience or knowledge to evaluate the interventions. This may discourage individuals’ continuous engagement in online health management. In this study, we are motivated to develop a personalized healthcare recommendation framework to help individuals better discover the interventions that fit their needs. Considering the challenges in intervention adaptation and diversification in a highly dynamic online healthcare environment, we propose an innovative online learning framework that synthesizes deep representation learning and a theory-guided diversity promotion scheme. We evaluate our approach through a real-world data set on users’ intervention participation in an online weight-loss community. Our results provide strong evidence for the effectiveness of our proposed recommendation framework and each of its design components. Our study contributes to the emerging information systems research on prescriptive analytics and the application of business intelligence. The proposed modeling framework and evaluation results offer important implications for multiple stakeholders, including online healthcare platforms, policymakers, and users. History: Ahmed Abbasi, Senior Editor; Huimin Zhao, Associate Editor. Funding: This work was supported in part by the National Natural Science Foundation of China [Grant 71729001]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2022.1191. Collapse

Topics: recommender system logistics management online learning system collaborative filtering usability

Methods: deep learning computational algorithm artificial neural network experiment design artifact

Theories: social cognitive theory

Using Forum and Search Data for Sales Prediction of High-Involvement Projects

2017 | Management Information Systems Quarterly | Citations: 0

Authors: Geva, Tomer; Oestreicher-Singer, Gal; Efron, Niv; Shimshoni, Yair

Abstract: A large body of research uses data from social media websites to predict offline ... Expand

Abstract: A large body of research uses data from social media websites to predict offline economic outcomes such as sales. However, recent research also points out that such data may be subject to various limitations and biases that may hurt predictive accuracy. At the same time, a growing body of research shows that a new source of online information, search engine logs, has the potential to predict offline outcomes. We study the relationship between these two important data sources in the context of sales predictions. Focusing on the automotive industry, a classic example of a domain of high-involvement products, we use Google’s comprehensive index of Internet discussion forums, in addition to Google search trend data. We find that adding search trend data to models based on the more commonly used social media data significantly improves predictive accuracy. We also find that predictive models based on inexpensive search trend data provide predictive accuracy that is comparable, at least, to that of social media data-based predictive models. Last, we show that the improvement in accuracy is considerably larger for “value” car brands, while for “premium” car brands the improvement obtained is more moderate. Keywords: Search trends, forums, social media, word-of-mouth, consumers’ interest, sales prediction, online data, high involvement products Collapse

Topics: social media Google+ internet technology website accounting

Methods: machine learning computational algorithm forums posting sentiment analysis robustness check

Do Customers Speak Their Minds? Using Forums and Search for Predicting Sales

2013 | International Conference on Information Systems | Citations: 4

Authors: Geva, Tomer; Oestreicher-Singer, Gal; Efron, Niv; Shimshoni, Yair

Abstract: A wide body of research uses data from social media websites to predict offline ... Expand

Abstract: A wide body of research uses data from social media websites to predict offline economic outcomes such as sales. However, in practice, such data are costly to collect and process. Additionally, sales forecasts based on social media data may be hampered by people’s tendency to restrict the topics they publicly discuss. Recently, a new source of predictive information—search engine logs—has become available. Interestingly, the relationship between these two important data sources has not been studied. Specifically, do they contain complementary information? Or does the information conveyed by one source render the information conveyed by the other source redundant? This study uses Google’s comprehensive index of internet discussion forums, in addition to Google search trend data. Predictive models based on search trend data are shown to outperform and complement forum-data-based models. Furthermore, the two sources display substantially different patterns of predictive capacity over time. Collapse

Topics: Google+ website internet technology price management search engine

Methods: computational algorithm forums posting machine learning descriptive statistic sentiment analysis