RecSys Issues Ontology: A Knowledge Classification of Issues for Recommender Systems Researchers

2020 | Information Systems Frontiers | Citations: 1

Authors: Bunnell, Lawrence; Osei-Bryson, Kweku-Muata; Yoon, Victoria Y.

Abstract: Scholarly research has extensively examined a number of issues and challenges af ... Expand

Abstract: Scholarly research has extensively examined a number of issues and challenges affecting recommender systems (e.g. ‘cold-start’, ‘scrutability’, ‘trust’, ‘context’, etc.). However, a comprehensive knowledge classification of the issues involved with recommender systems research has yet to be developed. A holistic knowledge representation of the issues affecting a domain is critical for research advancement. The aim of this study is to advance scholarly research within the domain of recommender systems through formal knowledge classification of issues and their relationships to one another within recommender systems research literature. In this study, we employ a rigorous ontology engineering process for development of a recommender system issues ontology. This ontology provides a formal specification of the issues affecting recommender systems research and development. The ontology answers such questions as, “What issues are associated with ‘trust’ in recommender systems research?”, “What are issues associated with improving and evaluating the ‘performance’ of a recommender system?” or “What ‘contextual’ factors might a recommender systems developer wish to consider in order to improve the relevancy and usefulness of recommendations?” Additionally, as an intermediate representation step in the ontology acquisition process, a concept map of recommender systems issues has been developed to provide conceptual visualization of the issues so that researchers may discern broad themes as well as relationships between concepts. These knowledge representations may aid future researchers wishing to take an integrated approach to addressing the challenges and limitations associated with current recommender systems research. Collapse

Topics: recommender system knowledge representation research and development organizational value privacy

Methods: ontological modelling qualitative content analysis machine learning domain ontology natural language processing

Detecting Anomalous Online Reviewers: An Unsupervised Approach Using Mixture Models

2019 | Journal of Management Information Systems | Citations: 48

Authors: Kumar, Naveen; Venugopal, Deepak; Qiu, Liangfei; Kumar, Subodha

Abstract: Online reviews play a significant role in influencing decisions made by users in ... Expand

Abstract: Online reviews play a significant role in influencing decisions made by users in day-to-day life. The presence of reviewers who deliberately post fake reviews for financial or other gains, however, negatively impacts both users and businesses. Unfortunately, automatically detecting such reviewers is a challenging problem since fake reviews do not seem out-of-place next to genuine reviews. In this paper, we present a fully unsupervised approach to detect anomalous behavior in online reviewers. We propose a novel hierarchical approach for this task in which we (1) derive distributions for key features that define reviewer behavior, and (2) combine these distributions into a finite mixture model. Our approach is highly generalizable and it allows us to seamlessly combine both univariate and multivariate distributions into a unified anomaly detection system. Most importantly, it requires no explicit labeling (spam/not spam) of the data. Our newly developed approach outperforms prior state-of-the-art unsupervised anomaly detection approaches. Collapse

Topics: social media email spam fake review deception detection online review

Methods: anomaly detection computational algorithm experiment machine learning unsupervised learning

TOWARDS AN INTEGRATIVE THEORETICAL FRAMEWORK OF INTERACTIVE MACHINE LEARNING SYSTEMS

2019 | European Conference On Information Systems | Citations: 0

Authors: MartÃŁnez, Miguel Angel Meza; Nadj, Mario; Maedche, Alexander

Abstract: Interactive machine learning (IML) is a learning process in which a user interac ... Expand

Abstract: Interactive machine learning (IML) is a learning process in which a user interacts with a system to iteratively define and optimise a model. Although recent years have illustrated the proliferation of IML systems in the fields of Human-Computer Interaction (HCI), Information Systems (IS), and Computer Science (CS), current research results are scattered leading to a lack of integration of existing work on IML. Furthermore, due to diverging functionalities and purposes IML systems can refer to, an uncertainty exists regarding the underlying distinct capabilities that constitute this class of systems. By reviewing extensive IML literature, this paper suggests an integrative theoretical framework for IML systems to address these current impediments. Reviewing 2,879 studies in leading journals and conferences during the years 1966-2018, we found an extensive range of applications areas that have implemented IML systems and the necessity to standardise the evaluation of those systems. Our framework offers an essential step to provide a theoretical foundation to integrate concepts and findings across different fields of research. The main contribution of this paper is organising and structuring the body of knowledge in IML for the advancement of the field. Furthermore, we suggest three opportunities for future IML research. From a practical point of view, our integrative theoretical framework can serve as a reference guide to inform the design and implementation of IML systems. Collapse

Topics: database system information retrieval human computer interaction usability information retrieval system

Methods: cluster analysis structured literature research cross sectional research reinforcement learning literature study

Design Principles for Systematic Search Systems: A Holistic Synthesis of a Rigorous Multi-cycle Design Science Research Journey

2019 | Business & Information Systems Engineering | Citations: 0

Authors: Sturm, Benjamin; Sunyaev, Ali

Abstract: Rigorous systematic literature searches are often described as complex, error-pr ... Expand

Abstract: Rigorous systematic literature searches are often described as complex, error-prone and time-consuming because of a prevailing lack of adequate technological assistance. Nonetheless, one of the first steps when conducting a rigorous literature review is finding an appropriate literature sample. The quality of this literature sample is an important factor for the overall quality of the literature review. This article investigates how to design innovative IT systems that effectively facilitate systematic literature searches. Applying the design science research paradigm, the research method consists of multiple design cycles of artifact development, evaluation, and refinement. In doing so, six design principles are derived that intend to increase the comprehensiveness, precision, and reproducibility of systematic literature searches. The results could be helpful for research and practice. The derived design knowledge builds a foundation for future research on systematic search systems and enables new methodological contributions. The results could also guide the development of innovative search systems and features that, eventually, increase the quality and efficiency of information accumulation in different contexts. Collapse

Topics: innovation management digital innovation information storage and retrieval system information retrieval information retrieval system

Methods: literature study design artifact design principle design science design requirement

Theories: theory of task technology fit

Implementing and Evaluating Serendipity in Delivering Personalized Health Information

2018 | ACM Transactions on Management Information Systems | Citations: 3

Authors: Fan, Xiangyu; Niu, Xi

Abstract: Serendipity has been recognized to have the potential of enhancing unexpected in ... Expand

Abstract: Serendipity has been recognized to have the potential of enhancing unexpected information discovery. This study shows that decomposing the concept of serendipity into unexpectedness and interest is a useful way for implementing this concept. Experts’ domain knowledge helps in providing serendipitous recommendation, which can be further improved by adaptively incorporating users’ real-time feedback. This research also conducts an empirical user-study to analyze the influence of serendipity in a health news delivery context. A personalized filtering system named MedSDFilter was developed, on top of which serendipitous recommendation was implemented using three approaches: random, static-knowledge-based, and adaptive-knowledge-based models. The three different models were compared. The results indicate that the adaptive-knowledge-based method has the highest ability in helping people discover unexpected and interesting contents. The insights of the research will make researchers and practitioners rethink the way in which search engines and recommender systems operate to address the challenges of discovering unexpected and interesting information. The outcome will have implications for empowering ordinary people with more chances of bumping into beneficial information. Collapse

Topics: healthcare data information retrieval system health information system recommender system IT productivity

Methods: parametric test one-way ANOVA survey design survey machine learning method

The Style of Tetris is…Possibly Tetris?: Creative Professionals’ Description of Video Game Visual Styles

2017 | HICSS | Citations: 0

Authors: Keating, Stephen; Lee, Wan-Chen; Windleharth, Travis; Lee, Jin Ha

Abstract: Despite the increasing importance of video games in both cultural and commercial ... Expand

Abstract: Despite the increasing importance of video games in both cultural and commercial aspects, typically they can only be accessed and browsed through limited metadata such as platform or genre. We explore visual styles of games as a complementary approach for providing access to games. In particular, we aimed to test and evaluate the existing visual style taxonomy developed in prior research with video game professionals and creatives. User data were collected from video game art and design students at the DigiPen Institute of Technology to gain insight into the relevance of the existing taxonomy to a professional audience. Using a think-aloud card sort method, we observed their thought process for describing and categorizing visual styles of video games, and also collected candidate terms for revising the taxonomy. The results of this research will inform ongoing metadata work in the field to develop a standard for cataloging video games and interactive media, and will be useful to information systems that sort and classify games for users and cultural preservation. Collapse

Topics: video game decision making software developer information retrieval system digital media

Methods: design science cognitive walkthrough card sorting usability testing qualitative coding

Training IBM Watson Using Automatically Generated Question-Answer Pairs

2017 | HICSS | Citations: 0

Authors: Lee, Jangho; Kim, Gyuwan; Yoo, Jaeyoon; Jung, Changwoo; Kim, Minseok; Yoon, Sungroh

Abstract: IBM Watson is a cognitive computing system capable of question answering in natu ... Expand

Abstract: IBM Watson is a cognitive computing system capable of question answering in natural languages. It is believed that IBM Watson can understand large corpora and answer relevant questions more effectively than any other question-answering system currently available. To unleash the full power of Watson, however, we need to train its instance with a large number of wellprepared question-answer pairs. Obviously, manually generating such pairs in a large quantity is prohibitively time consuming and significantly limits the efficiency of Watson’s training. Recently, a large-scale dataset of over 30 million question-answer pairs was reported. Under the assumption that using such an automatically generated dataset could relieve the burden of manual question-answer generation, we tried to use this dataset to train an instance of Watson and checked the training efficiency and accuracy. According to our experiments, using this auto-generated dataset was effective for training Watson, complementing manually crafted question-answer pairs. To the best of the authors’ knowledge, this work is the first attempt to use a largescale dataset of automatically generated questionanswer pairs for training IBM Watson. We anticipate that the insights and lessons obtained from our experiments will be useful for researchers who want to expedite Watson training leveraged by automatically generated question-answer pairs. Collapse

Topics: Wikipedia competitive advantage knowledge base online learning system website

Methods: experiment machine learning literature study cross sectional research

Discovering Finance Keywords via Continuous-Space Language Models

2016 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Tsai, Ming-Feng; Wang, Chuan-Ju; Chien, Po-Chuan

Abstract: The growing amount of public financial data makes it increasingly important to l ... Expand

Abstract: The growing amount of public financial data makes it increasingly important to learn how to discover valuable information for financial decision making. This article proposes an approach to discovering financial keywords from a large number of financial reports. In particular, we apply the continuous bag-of-words (CBOW) model, a well-known continuous-space language model, to the textual information in 10-K financial reports to discover new finance keywords. In order to capture word meanings to better locate financial terms, we also present a novel technique to incorporate syntactic information into the CBOW model. Experimental results on four prediction tasks using the discovered keywords demonstrate that our approach is effective for discovering predictability keywords for post-event volatility, stock volatility, abnormal trading volume, and excess return predictions. We also analyze the discovered keywords that attest to the ability of the proposed method to capture both syntactic and contextual information between words. This shows the success of this method when applied to the field of finance. Collapse

Topics: database system data breach accounting information retrieval system IT career

Methods: part of speech tagging experiment financial report language model word2vec

Fully automated resource retrieval in telecommunications and internet converged environments

2014 | Information Systems Frontiers | Citations: 1

Authors: Suarez-Meza, Luis Javier; Zúñiga, Julián Andrés; Pedraza, Edgar Camilo; Corrales, Juan Carlos

Abstract: Today, most of the current retrieval approaches are targeted only to Web service ... Expand

Abstract: Today, most of the current retrieval approaches are targeted only to Web services, leaving out different other resource types (Telecom, IT and content). Moreover, such approaches are not aimed at ordinary users, since to find and use a service, users must specify their requests by complex expressions that represent desired services. Thus, the search and selection of relevant resources depend on the ability of the user to retrieve the most suitable ones, which is an inefficient task. In order to address these issues, in this paper we propose an innovative approach for resource retrieval over Telecommunications and Internet converged environments, which brings together traditional NLP techniques with retrieval process based on lightweight semantic approachs. The preliminary experiments show promising results in contrast to traditional approaches. Collapse

Topics: communication service infrastructure knowledge representation internet technology web service Flickr

Methods: natural language processing WordNet computational algorithm experiment machine learning

An adaptive learning to rank algorithm: Learning automata approach

2012 | Decision Support Systems | Citations: 0

Authors: Akbari Torkestani, Javad

Abstract: The recent years have witnessed the birth and explosive growth of the web. It is ... Expand

Abstract: The recent years have witnessed the birth and explosive growth of the web. It is obvious that the exponential growth of the web has made it into a huge interconnected source of information wherein finding a document without a searching tool is unimaginable. Today's search engines try to provide the most relevant suggestions to the user queries. To do this, different strategies are used to enhance the precision of the information retrieval process. In this paper, a learning method is proposed to rank the web documents in a search engine. The proposed method takes advantage of the user feedback to enhance the precision of the search results. To do so, it uses a learning automata-based approach to train the search engine. In this method, the user feedback is defined as its interest to review an item. Within the search results, the document that is visited by the user is more likely relevant to the user query. Therefore, its choice probability must be increased by the learning automaton. By this, the rank of the most relevant documents increases as that of the others decreases. To investigate the efficiency of the proposed method, extensive simulation experiment is conducted on well-known data collections. The obtained results show the superiority of the proposed approach over the existing methods in terms of mean average precision, precision at position n, and normalized discount cumulative gain. Collapse

Topics: search engine information retrieval system information retrieval

Methods: computational algorithm experiment simulation machine learning binary classification