Results

Found 14 articles

Clear filters

Benchmarking the Robustness of Phishing Email Detection Systems

2023 | Americas Conference on Information Systems | Citations: 0
Authors: Ampel, Benjamin; Gao, Yang; Hu, James; Samtani, Sagar; Chen, Hsinchun

Abstract: Social engineering attacks are currently the most cited cybersecurity threat to ... Expand

Abstract: Social engineering attacks are currently the most cited cybersecurity threat to organizations. Phishing emails are the most salient form of social engineering attacks. Organizations are increasingly implementing AI-enabled systems to detect phishing emails. However, AI-enabled systems are often susceptible to textual perturbations, where an adversary makes a small change to cause a misclassification. In this study, we sought to identify the performance of prevailing phishing email detection systems (PEDS) against character, word, sentence, and multi-level adversarial text perturbations. Through a principled benchmarking framework, we quantitatively demonstrated the lack of robustness prevailing PEDS have to specific types of text-based adversarial perturbations (e.g., character, word, sentence, multi-level). The results of this study provide new insights into the robustness of AI-based PEDS and highlight the need for organizations to adopt a multi-layered approach to phishing protection. Additionally, organizations can implement our benchmark framework to test their PEDS against adversarial perturbations. Collapse

Semantic filters: pre-trained language model

Topics: artificial intelligence electronic mail email phishing phishing IT security

Methods: word2vec random forest classification naïve bayes literature study pre-trained language model
Automated Deductive Content Analysis of Text: A Deep Contrastive and Active Learning Based Approach

2023 | International Conference on Information Systems | Citations: 0
Authors: De Oliveira Silveira, Alysson; Ray, Arindam; Ebrahimi, Mohammadreza; Bhattacherjee, Anol

Abstract: Content analysis traditionally involves human coders manually combing through t ... Expand

Abstract: Content analysis traditionally involves human coders manually combing through text documents to search for relevant concepts and categories. However, this approach is time-intensive and not scalable, particularly for secondary data like social media content, news articles, or corporate reports. To address this problem, the paper presents an automated framework called Automated Deductive Content Analysis of Text (ADCAT) that uses deep learning-based semantic techniques, an ontology of validated construct measures, a large language model, human-in-the-loop disambiguation, and a novel augmentation-based weighted contrastive learning approach for improved language representations, to build a scalable approach for deductive content analysis. We demonstrate the effectiveness of the proposed approach to identify firm innovation strategies from their 10-K reports to obtain inferences reasonably close to human coding. Collapse

Semantic filters: pre-trained language model

Topics: knowledge representation human in the loop human computer interaction innovation management organizational innovation

Methods: ontological modelling machine learning language model large language model qualitative content analysis
Incorporating Multiple Knowledge Sources for Targeted Aspect-based Financial Sentiment Analysis
PDF
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

Semantic filters: pre-trained language model

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

Methods: sentiment analysis RoBERTa transformer model BERT XLNet
Detecting Feature Requests of Third-Party Developers through Machine Learning: A Case Study of the SAP Community

2023 | HICSS | Citations: 0
Authors: Kauschinger, M.; Vieth, N.; Schreieck, M.; Krcmar, H.

Abstract: The elicitation of requirements is central for the development of successful so ... Expand

Abstract: The elicitation of requirements is central for the development of successful software products. While traditional requirement elicitation techniques such as user interviews are highly labor-intensive, data-driven elicitation techniques promise enhanced scalability through the exploitation of new data sources like app store reviews or social media posts. For enterprise software vendors, requirements elicitation remains challenging because app store reviews are scarce and vendors have no direct access to users. Against this background, we investigate whether enterprise software vendors can elicit requirements from their sponsored developer communities through data-driven techniques. Following the design science methodology, we collected data from the SAP Community and developed a supervised machine learning classifier, which automatically detects feature requests of third-party developers. Based on a manually labeled data set of 1,500 questions, our classifier reached a high accuracy of 0.819. Our findings reveal that supervised machine learning models are an effective means for the identification of feature requests. Collapse

Semantic filters: pre-trained language model

Topics: enterprise information system requirements elicitation requirements analysis Fiverr cloud computing

Methods: machine learning supervised learning design science naïve bayes classification method
What Symptoms and How Long? An Interpretable AI Approach for Depression Detection in Social Media

2023 | International Conference on Information Systems | Citations: 0
Authors: Kuang, Junwei; Xie, Jiaheng; Yan, Zhijun

Abstract: Depression is the most prevalent and serious mental illness, which induces grav ... Expand

Abstract: Depression is the most prevalent and serious mental illness, which induces grave financial and societal ramifications. Depression detection is key for early intervention to mitigate those consequences. Such a high-stake decision inherently necessitates interpretability. Although a few depression detection studies attempt to explain the decision, these explanations misalign with the clinical depression diagnosis criterion that is based on depressive symptoms. To fill this gap, we develop a novel Multi-Scale Temporal Prototype Network (MSTPNet). MSTPNet innovatively detects and interprets depressive symptoms as well as how long they last. Extensive empirical analyses show that MSTPNet outperforms state-of-the-art depression detection methods. This result also reveals new symptoms that are unnoted in the survey approach. We further conduct a user study to demonstrate its superiority over the benchmarks in interpretability. This study contributes to IS literature with a novel interpretable deep learning model for depression detection in social media. Collapse

Semantic filters: pre-trained language model

Topics: social media healthcare analytics

Methods: design artifact deep learning computational algorithm machine learning ablation analysis
A deep learning approach for detecting fake reviewers: Exploiting reviewing behavior and textual information
PDF
2023 | Decision Support Systems | Citations: 0
Authors: Zhang, Dong; Li, Wenwen; Niu, Baozhuang; Wu, Chong

Abstract: Ensuring the credibility of online consumer reviews (OCRs) is a growing societal ... Expand

Abstract: Ensuring the credibility of online consumer reviews (OCRs) is a growing societal concern. However, the problem of fake reviewers on online platforms significantly influences e-commerce authenticity and consumer trust. Existing studies for fake reviewer detection mainly focus on deriving novel behavioral and linguistic features. These features require extensive human labor and expertise, placing a heavy burden on platforms. Therefore, we propose a novel end-to-end framework to detect fake reviewers based on behavior and textual information. It has two key components: (1) a behavior-sensitive feature extractor that learns the underlying patterns of reviewing behavior; (2) a context-aware attention mechanism that extracts valuable features from online reviews. We rigorously evaluate each proposed module and the entire framework against state-of-the-art benchmarks on two real-world datasets from http://Yelp.com. Experimental results demonstrate that our method achieves state-ofthe-art results on fake reviewer detection. Our method can be considered a tentative step toward lowering human labor costs in realizing automated fake reviewer detection on e-commerce platforms. Collapse

Semantic filters: pre-trained language model

Topics: fake review online review electronic commerce deception detection shapley additive explanation

Methods: machine learning digital trace data experiment parametric test deep learning
Towards a Generation of Artificially Intelligent Strategy Tools: The SWOT Bot

2022 | European Conference On Information Systems | Citations: 0
Authors: Au, Christian; Winkler, Till; Paul, Herbert

Abstract: Strategy tools are widely used to inform the complex and unstructured decision ... Expand

Abstract: Strategy tools are widely used to inform the complex and unstructured decision making of firms. Although software has evolved to support strategy analysis, such digital strategy tools still require heavy manual work especially on the data input and processing levels, making their use time-intensive, costly, and susceptible to biases. This design research presents the 'SWOT Bot', a digital strategy tool that exploits recent advances in natural language processing (NLP) to perform a SWOT (strengths, weaknesses, opportunities, threats) analysis. Our artifact uses a feed reader, an NLP pipeline, and a visual interface to automatically extract information from a text corpus (e.g., analyst reports) and present it to the user. We argue that the SWOT Bot reduces time and adds objectivity to strategy analyses while allowing the human-in-the-loop to focus on value-adding tasks. Besides providing a functioning prototype, our work provides three general design principles for the development of next-generation digital strategy tools.Strategy tools-such as a SWOT, Porter's Five Forces, or the PEST model-are widely used among managers, analysts, and consultants in today's business practice (e.g., Jarzabkowski and Kaplan, 2014;Wright et al., 2013;Vaara andWhittington, 2012 andor Clark, 1997). While the term 'tool' generally refers to anything that helps perform a specific task, management tools cover any model, concept, framework, method, or technique that helps to solve a managerial problem (Hakala and Vuorinen, 2020; Jarzabkowski and Kaplan, 2014;Knott, 2008). Strategy tools are a subset of management tools that deal with strategic challenges, such as understanding a company's positioning and developing a new competitive strategy. Strategy tools are cognitive as well as material artifacts (Paroutis et al., 2015) that shape the mental models of strategists and hence affect both content and process of strategy work (Vuorinen et al., 2018). The application of strategy tools in practice is based on quantitative and qualitative inputs, where qualitative information seems to dominate. Although the use of strategy tools has evolved from pen and paper drawings to digital templates and dedicated software suites over the past decades (Ain et al., 2019;Arnott et al., 2017), our initial analysis found that digital tools for strategy available on the market lack support on the input, processing, and output levels. Contemporary digital strategy tools heavily rely on human tasks because strategic analysis is a complex process that integrates knowledge and data from heterogeneous and often unstructured sources (Hakala and Vuorinen, 2020;Jarzabkowski and Kaplan, 2015). While human involvement in strategy processes matters (Reeves and Ueda, 2016), it also comes with certain costs (Grant, 2016). Collapse

Semantic filters: pre-trained language model

Topics: information systems strategy application programming interface Google+ SWOT analysis usability

Methods: natural language processing design artifact design science design requirement BERT
Comparative Analysis of Classical and Deep Learning-based Natural Language Processing for Prioritizing Customer Complaints
PDF
2022 | HICSS | Citations: 0
Authors: Blümel, Jan; Zaki, Mohamed

Abstract: Recent advancements in natural language processing have been shown to be very ef ... Expand

Abstract: Recent advancements in natural language processing have been shown to be very effective for different text mining tasks and thus have provided the opportunity to enhance service research. To improve the customer service experience, this paper compares several natural language processing approaches in order to automatically prioritize incoming customer complaints for service agents. This can help companies to reduce customers’ friction and enable effective resource allocations. Our paper uses stateof-the-art feature engineering techniques (e.g., term frequency, TF-IDF and Word2Vec) to identify key words that could enable machine to prioritize complainers. We experimented with many classical machine learning classification algorithms, such as Random Forests, Support Vector Machines, Decision Trees and Logistic Regression, as well as with deep learning-based classifiers, such as convolutional neural networks, bidirectional long short-term memory, and the pre-trained language model BERT to compare the model performance. Our findings show that the pre-trained language model BERT and TFIDF in combination with Logistic Regression yields the highest macro averaged F1-score across the multiple classes and is therefore most capable of predicting the priority group of incoming customer complaints. Collapse

Semantic filters: pre-trained language model

Topics: Wikipedia customer service database system Google+ application programming interface

Methods: natural language processing deep learning machine learning term frequency–inverse document frequency word2vec
Where was COVID-19 first discovered? Designing a question-answering system for pandemic situations

2022 | European Conference On Information Systems | Citations: 0
Authors: Graf, Johannes; Lancho, Gino; Zschech, Patrick; Heinrich, Kai

Abstract: The COVID-19 pandemic is accompanied by a massive "infodemic" that makes it har ... Expand

Abstract: The COVID-19 pandemic is accompanied by a massive "infodemic" that makes it hard to identify concise and credible information for COVID-19-related questions, like incubation time, infection rates, or the effectiveness of vaccines. As a novel solution, our paper is concerned with designing a question-answering system based on modern technologies from natural language processing to overcome information overload and misinformation in pandemic situations. To carry out our research, we followed a design science research approach and applied Ingwersen's cognitive model of information retrieval interaction to inform our design process from a socio-technical lens. On this basis, we derived prescriptive design knowledge in terms of design requirements and design principles, which we translated into the construction of a prototypical instantiation. Our implementation is based on the comprehensive CORD-19 dataset, and we demonstrate our artifact's usefulness by evaluating its answer quality based on a sample of COVID-19 questions labeled by biomedical experts. Collapse

Semantic filters: pre-trained language model

Topics: knowledge base pandemic covid 19 database system Elasticsearch

Methods: design artifact design principle design science question answering design requirement
Transformer-based Summarization and Sentiment Analysis of SEC 10-K Annual Reports for Company Performance Prediction
PDF
2022 | HICSS | Citations: 0
Authors: Hsieh, Hsin-Ting; Hristova, Diana

Abstract: Annual reports published by companies contain important insights regarding their ... Expand

Abstract: Annual reports published by companies contain important insights regarding their performance and are often analyzed in a manual, subjective manner. We address this point by combining the streams of research on text summarization and topic modelling with the one on sentiment analysis. Our approach consists of the steps of text summarization using BERTSUMEXT, topic modelling with LDA, sentiment analysis with FinBERT, and performance prediction with Decision Trees and Random Forest. The result provides decision makers with an interpretable and condensed representation of the content of annual reports, together with its relationship to future company performance. We evaluate our approach on 10-K reports, demonstrating both its interpretability for analysts and explanatory power regarding future company performance. Collapse

Semantic filters: pre-trained language model

Topics: organizational productivity data breach cloud computing Python explainable artificial intelligence

Methods: sentiment analysis topic model machine learning BERT decision tree classification

Page 1 of 2 Next

AIS11

Journals

Conferences

Semantic Filter

Found 14 articles

Benchmarking the Robustness of Phishing Email Detection Systems

Automated Deductive Content Analysis of Text: A Deep Contrastive and Active Learning Based Approach

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

Detecting Feature Requests of Third-Party Developers through Machine Learning: A Case Study of the SAP Community

What Symptoms and How Long? An Interpretable AI Approach for Depression Detection in Social Media

A deep learning approach for detecting fake reviewers: Exploiting reviewing behavior and textual information

Towards a Generation of Artificially Intelligent Strategy Tools: The SWOT Bot

Comparative Analysis of Classical and Deep Learning-based Natural Language Processing for Prioritizing Customer Complaints

Where was COVID-19 first discovered? Designing a question-answering system for pandemic situations

Transformer-based Summarization and Sentiment Analysis of SEC 10-K Annual Reports for Company Performance Prediction