Incorporating Multiple Knowledge Sources for Targeted Aspect-based Financial Sentiment Analysis

2023 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Du, Kelvin; Xing, Frank; Cambria, Erik

Abstract: Combining symbolic and subsymbolic methods has become a promising strategy as re ... Expand

Abstract: Combining symbolic and subsymbolic methods has become a promising strategy as research tasks in AI grow increasingly complicated and require higher levels of understanding. Targeted Aspect-based Financial Sentiment Analysis (TABFSA) is an example of such complicated tasks, as it involves processes like information extraction, information specification, and domain adaptation. However, little is known about the design principles of such hybrid models leveraging external lexical knowledge. To fill this gap, we define anterior, parallel, and posterior knowledge integration and propose incorporating multiple lexical knowledge sources strategically into the fine-tuning process of pre-trained transformer models for TABFSA. Experiments on the Financial Opinion mining and Question Answering challenge (FiQA) Task 1 and SemEval 2017 Task 5 datasets show that the knowledge-enabled models systematically improve upon their plain deep learning counterparts, and some outperform state-of-the-art results reported in terms of aspect sentiment analysis error. We discover that parallel knowledge integration is the most effective and domain-specific lexical knowledge is more important according to our ablation analysis. Collapse

Topics: microblog domain-specific language Twitter social media knowledge graph

Methods: sentiment analysis RoBERTa transformer model BERT XLNet

Webcam Eye Tracking for Desktop and Mobile Devices: A Systematic Review

2023 | HICSS | Citations: 0

Authors: Heck, Melanie; Becker, Christian; Deutscher, Viola

Abstract: Building the Internet of Behaviors (IOB) obviously requires capturing human beha ... Expand

Abstract: Building the Internet of Behaviors (IOB) obviously requires capturing human behavior. Sensor input from eye tracking has been widely used for profiling in market research, adaptive user interfaces, and other smart systems, but requires dedicated hardware. The wide spread of webcams in consumer devices like phones, tablets, notebooks, and smart TVs has fostered eye tracking with commodity cameras. In this paper, we present a systematic review across the IEEE and ACM databases -- complemented by snowballing and input from eye tracking experts at CHI 2021 -- to list and characterize publicly available webcam eye trackers that estimate the point-of-regard on devices with no additional hardware. Information from articles was supplemented by searching author websites and code repositories, and contacting authors. 16 eye trackers were found that can be used. The restrictions regarding license terms and technical performance are presented, enabling developers to choose an appropriate software for their IoB application. Collapse

Topics: tablet source code database system desktop computer website

Methods: oculometry literature sample structured literature research meta analysis synthesis

Guest Editorial: Big data-driven theory building: Philosophies, guiding principles, and common traps

2023 | International Journal of Information Management | Citations: 0

Authors: Kar, Arpan Kumar; Angelopoulos, Spyros; Rao, H. Raghav

Abstract: While data availability and access used to be a major challenge for information ... Expand

Abstract: While data availability and access used to be a major challenge for information systems research, the growth and ease of access to large datasets and data analysis tools has increased interest to use such resources for publishing. Such publications, however, seem to offer weak theoretical contributions. While big data-driven studies increasingly gain popularity, they rarely introspect why a phenomenon is better explained by a theory and limit the analysis to data descriptive by mining and visualizing large volumes of big data. We address this pressing need and provide directions to move towards theory building with Big Data. We differentiate based on inductive and deductive approaches and provide guidelines how may undertake steps for theory building. In doing so, we further provide directions surrounding common pitfalls that should be avoided in this journey of Big-Data driven theory building. Collapse

Topics: big data social media misinformation online community user experience

Methods: theory development robustness check natural language processing exploratory data analysis artificial neural network

Theories: transaction cost economics

Facilitators and Barriers of Artificial Intelligence Adoption in Business – Insights from Opinions Using Big Data Analytics

2023 | Information Systems Frontiers | Citations: 0

Authors: Kar, Arpan Kumar; Kushwaha, Amit Kumar

Abstract: Data-driven predictions have become an inseparable part of business decisions. A ... Expand

Abstract: Data-driven predictions have become an inseparable part of business decisions. Artificial Intelligence (AI) has started helping the product and support teams perform more accurate experiments in various business settings. This study proposes a framework for businesses based on inductive learnings related to success and barriers shared on social media platforms. Our goal is to analyse the signals emerging from these conversational opinions from the early adoption of AI, with a focus towards facilitators and barriers faced by teams. Factors like efficiency, innovation, business research, product novelty, manual intervention, adaptability, emotion, support, personal growth, experiential learning, fear of failure and fear of upgradation have been identified based on an exploratory study and then a confirmatory study. We present the learnings through a roadmap for practitioners. This study contributes to the IS literature by delineating AI as a determinant of success and introduces a lot of organizational factors into the model. Collapse

Topics: artificial intelligence Twitter innovation management big data social media

Methods: topic model literature study synthesis regression analysis method data transformation

Theories: organizational theory organizational behavior theory management theory technology acceptance model organizational learning theory

Diffusion of real versus misinformation during a crisis event: A big data-driven approach

2023 | International Journal of Information Management | Citations: 0

Authors: King, Kelvin K.; Wang, Bin

Abstract: Misinformation has captured the interest of academia in recent years with severa ... Expand

Abstract: Misinformation has captured the interest of academia in recent years with several studies looking at the topic broadly with inconsistent results. In this research, we attempt to bridge the gap in the literature by examining the impacts of user-, time-, and content-based characteristics that affect the virality of real versus misinformation during a crisis event. Using a big data-driven approach, we collected over 42 million tweets during Hurricane Harvey and obtained 3589 original verified real or false tweets by cross-checking with fact-checking websites and a relevant federal agency. Our results show that virality is higher for misinformation, novel tweets, and tweets with negative sentiment or lower lexical density. In addition, we reveal the opposite impacts of sentiment on the virality of real news versus misinformation. We also find that tweets on the environment are less likely to go viral than the baseline religious news, while real social news tweets are more likely to go viral than misin­ formation on social news. Collapse

Topics: Twitter misinformation social media hurricane big data

Methods: machine learning computational algorithm natural language processing decision tree classification gradient boosting

An AI-based Decision Support System for Predicting Mental Health Disorders

2023 | Information Systems Frontiers | Citations: 0

Authors: Tutun, Salih; Johnson, Marina E.; Ahmed, Abdulaziz; Albizri, Abdullah; Irgil, Sedat; Yesilkaya, Ilker; Ucar, Esma Nur; Sengun, Tanalp; Harfouche, Antoine

Abstract: Approximately one billion individuals suffer from mental health disorders, such ... Expand

Abstract: Approximately one billion individuals suffer from mental health disorders, such as depression, bipolar disorder, schizophrenia, and anxiety. Mental health professionals use various assessment tools to detect and diagnose these disorders. However, these tools are complex, contain an excessive number of questions, and require a significant amount of time to administer, leading to low participation and completion rates. Additionally, the results obtained from these tools must be analyzed and interpreted manually by mental health professionals, which may yield inaccurate diagnoses. To this extent, this research utilizes advanced analytics and artificial intelligence to develop a decision support system (DSS) that can efficiently detect and diagnose various mental disorders. As part of the DSS development process, the Network Pattern Recognition (NEPAR) algorithm is first utilized to build the assessment tool and identify the questions that participants need to answer. Then, various machine learning models are trained using participants’ answers to these questions and other historical data as inputs to predict the existence and the type of their mental disorder. The results show that the proposed DSS can automatically diagnose mental disorders using only 28 questions without any human input, to an accuracy level of 89%. Furthermore, the proposed mental disorder diagnostic tool has significantly fewer questions than its counterparts; hence, it provides higher participation and completion rates. Therefore, mental health professionals can use this proposed DSS and its accompanying assessment tool for improved clinical decision-making and diagnostic accuracy. Collapse

Topics: artificial intelligence decision support system explainable artificial intelligence information portal decision making

Methods: computational algorithm support vector machine design process lasso regression ridge regression

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

A machine learning investigation of factors that contribute to predicting cognitive performance: Difficulty level, reaction time and eye-movements

2022 | Decision Support Systems | Citations: 0

Authors: Bachurina, Valentina; Sushchinskaya, Svetlana; Sharaev, Maxim; Burnaev, Evgeny; Arsalidou, Marie

Abstract: Predicting accuracy in cognitively challenging tasks has potential applications ... Expand

Abstract: Predicting accuracy in cognitively challenging tasks has potential applications in education and industry. Task demand has been linked with increases in response time and variations in reaction time and eye-tracking metrics, however, machine learning research has not been used to predict performance on tasks with multiple levels of difficulty. We report data on adult participants who performed tasks of mental attentional capacity with six levels of difficulty and use machine learning methods to predict accuracy scores considering metrics associated with task difficulty, reaction time and eye movements. Results show that machine learning models can robustly predict performance with reaction times and difficulty level being the strongest predictors. Eye-tracking indices can also predict accuracy independently, with the most important features of the model driven by the number of fixations, number of saccades, duration of the current fixation and pupil size. Practical and theoretical impli­ cations of the results are discussed. Collapse

Topics: Python decision support system software developer

Methods: oculometry machine learning descriptive statistic computational algorithm regression analysis method

Theories: cognitive theory

Learning Faces to Predict Matching Probability in an Online Matching Platform

2022 | International Conference on Information Systems | Citations: 0

Authors: Kwon, Soonjae; Park, Sunghyuk; Lee, Gene Moo; Lee, Dongwon

Abstract: With the increasing use of online matching platforms, predicting matching proba ... Expand

Abstract: With the increasing use of online matching platforms, predicting matching probability between users is crucial for efficient market design. Although previous studies have constructed various visual features to predict matching probability, facial features, which are important in online matching, have not been widely used. We find that deep learning-enabled facial features can significantly enhance the prediction accuracy of a user's partner preferences from the individual rating prediction analysis in an online dating market. We also build prediction models for each gender and use prior theories to explain different contributing factors of the models. Furthermore, we propose a novel method to visually interpret facial features using the generative adversarial network (GAN). Our work contributes the literature by providing a framework to develop and interpret facial features to investigate underlying mechanisms in online matching markets. Moreover, matching platforms can predict matching probability more accurately for better market design and recommender systems. Collapse

Topics: online dating shapley additive explanation online labor market Google+ user behavior

Methods: StyleGAN deep learning generative adversarial network grounded theory artificial neural network

Product sampling on reward-based crowdfunding platforms

2022 | Information & Management | Citations: 0

Authors: Liu, Xiaochen; Fang, Yulin; Ye, Qiang

Abstract: On reward-based crowdfunding platforms, uncertainty in investment outcome severe ... Expand

Abstract: On reward-based crowdfunding platforms, uncertainty in investment outcome severely hinders transactions between online backers and fundraisers. The industry is dedicated to alleviating these dilemmas through campaign design and platform function design. In recent years, online product sampling has been introduced as a promising promotion instrument for fundraisers. Using a unique dataset of 4,027 campaign-level observations collected from China’s highly influential JD Crowdfunding website z.jd.com, this study discusses whether and to what extent a product sampling strategy is beneficial to crowdfunding campaigns. Differences in the strategy’s effectiveness across product types are also examined. Empirical results suggest that adopting online product sampling is helpful to crowdfunding performance because it increases the total number of backers, the total fundraising amounts, the average amount of funds raised per backer, the fundraising percentage, and the pos­ sibility of campaign success. The strength of product sampling, however, decreases along the search-experience product continuum. Collapse

Topics: crowdsourcing crowdfunding project productivity website price management

Methods: regression analysis method experimental group cross sectional research logistic regression proportional hazards model