Journal of the Association for Information Science and Technology最新文献

We argue that advances in large language models (LLMs) and generative Artificial Intelligence (AI) will diminish the value of Wikipedia, due to a withdrawal by human content producers, who will withhold their efforts, perceiving less need for their efforts and increased “AI competition.” We believe the greatest threat to Wikipedia stems from the fact that Wikipedia is a user-generated product, relying on the “selfish altruism” of its human contributors. Contributors who reduce their contribution efforts as AI pervades the platform, will thus leave Wikipedia increasingly dependent on additional AI activity. This, combined with a dynamic where readership creates authorship and readers being disintermediated, will inevitably cause a vicious cycle leading to a staling of the content and diminishing value of this venerable knowledge resource.

This brief communication presents a novel adaptation of common bibliometric measures to provide a quantitative assessment of an artist's music catalog that incorporates both impact and productivity. Data from Billboard's weekly Hot 100™ music charts are used to rank the all-time greatest artists. Since the sorted data are increasing in value—that is, a number 1 hit is best—a transformation is applied to provide a convex, monotonically decreasing curve. Furthermore, since conventional bibliometrics result in several artists with identical measures, metrics inspired by the multidimensional $��h�$- and $��g�$-indices are used to rank the artists. We find that this approach provides a simple, yet unbiased, approach for ranking the all-time greatest music artists.

When undertaking complex search scenarios, the underlying information need cannot be satisfied by finding a single optimal resource; instead, searchers need to engage in exploratory search processes to find multiple resources by iteratively revising and reformulation their queries. This process of query refinement is particularly challenging when using a mobile device, where typing is difficult. Furthermore, in mobile search contexts interruptions can lead to searchers losing track of what they were doing. To address these challenges, we designed a public digital library search interface for mobile devices that includes two novel features: drag-and-drop query refinement and query history visualization. To assess the value of this interface compared to a typical baseline, we conducted a controlled laboratory study with 32 participants that included pursuing complex search scenarios, being interrupted in the midst of the search, and resuming the search after the interruption. While participants took more time, they generated longer queries and reported positive subjective opinions about the usability of the exploratory search and task resumption features, along with a greater increase in certainty. These findings show the value of leveraging new touch-based interaction mechanisms within mobile search contexts, and the benefits that visualization can bring to supporting search task resumption.

Prior research in human–computer interaction suggests that system-level clarifications are necessary for understanding user intent and communicating effectively with the user. Such clarifications or explanations could contain the system's abstract knowledge of the search or a functional description of the search process (queries and information sources employed). While these interactions may aid the user and the agent in better understanding each other, very few studies have explored the influence of such clarifications on the users' search experience. This research examines whether and how system-level clarifications (or explanations) affect the user experience when searching through spoken dialogues. We analyzed user satisfaction and preferences in systems with and without explicit clarifications in a within-subjects Wizard-of-Oz user study. We recruited 25 participants and collected user–system interaction data for 50 search sessions. The user feedback was collected using pre- and post-task surveys and exit interviews. Statistical and qualitative analysis of user responses yielded some interesting findings. While Wilcoxon Signed Rank Test found that using explicit system-level clarifications had no positive influence on the user's search experience, the overall search experience degraded with system clarifications (Z = −2.066, p = 0.04). The user interview data provided valuable insights into how and when clarifications should be offered to the user.

The study explored college students' use of generative artificial intelligence (GenAI) tools, such as ChatGPT, for academic tasks and their perceptions and behaviors in assessing the credibility of GenAI-generated information. Semistructured interviews were conducted with 25 college students in the United States. Interview transcripts were analyzed using the qualitative content analysis method. The study identified various types of academic tasks for which students used ChatGPT, including writing, programming, and learning. Guided by two models of credibility assessment Hilligoss and Rieh (2008); Metzger (2007), six factors influencing students' motivation and ability to assess the credibility of GenAI-generated information were identified (e.g., task salience, social pressure). We also identified 9 constructs (e.g., refinedness, explainability), 5 heuristics (e.g., inter- and intrasystem consistency heuristics), and 10 cues (e.g., version and tone) used by students to assess the credibility of GenAI-generated information. This study provides theoretical and empirical findings regarding students' use of GenAI tools in the academic context and credibility evaluation of the system outputs using rich, qualitative interview data.

Communicating insights from data effectively requires design skills, technical knowledge, and experience. Data must be accurately represented with aesthetically pleasing visuals and engaging text to effectively communicate to the intended audience. Data storytelling has received much attention lately, but as of yet, it does not have a theoretical and practical foundation in information science. A data story adds context, narrative, and structure to the visual representation of data, providing audiences with character, plot, and a holistic experience of narrative. This paper proposes a methodological approach to transform a data visualization into a data story based on the Data-Information-Knowledge-Wisdom (DIKW) pyramid and the S-DIKW Framework. Starting from the bottom of the pyramid, the proposed approach defines a strategy to represent insights extracted from data. Data is then turned into information by identifying character(s) facing a problem, adding textual and graphic content; information is turned into knowledge by organizing what happens as a plot. Finally, a call to wise action—always informed by cultural and community values—completes the storytelling transformation to create a data story. This article contributes to the theoretical understanding of data stories as emerging information forms, supporting richer understandings of a story as information in the information sciences.

This review paper explores the evolution of discussions about “long-tail” scientific data in the scholarly literature. The “long-tail” concept, originally used to explain trends in digital consumer goods, was first applied to scientific data in 2007 to refer to a vast array of smaller, heterogeneous data collections that cumulatively represent a substantial portion of scientific knowledge. However, these datasets, often referred to as “long-tail data,” are frequently mismanaged or overlooked due to inadequate data management practices and institutional support. This paper examines the changing landscape of discussions about long-tail data over time, situated within broader ecosystems of research data management and the natural interplay between “big” and “small” data. The review also bridges discussions on data curation in Library & Information Science (LIS) and domain-specific contexts, contributing to a more comprehensive understanding of the long-tail concept's utility for effective data management outcomes. The review aims to provide a more comprehensive understanding of this concept, its terminological diversity in the literature, and its utility for guiding data management, overall informing current and future information science research and practice.

When using retrieval-augmented generation (RAG) to handle multi-document question answering (MDQA) tasks, it is beneficial to decompose complex queries into multiple simpler ones to enhance retrieval results. However, previous strategies always employ a one-shot approach of question decomposition, overlooking subquestions dependency problem and failing to ensure that the derived subqueries are single-hop. To overcome this challenge, we introduce a novel framework called DSRC-QCS. Decompose-solve-renewal-cycle (DSRC) is an iterative multi-hop question processing module. The key idea of DSRC involves using a unique symbol to achieve hierarchical dependency management and employing a cyclical process of question decomposition, solving, and renewal to continuously generate and resolve all single-hop subquestions. Query-chain selector (QCS) functions as a voting mechanism that effectively utilizes the reasoning process of DSRC to assess and select solutions. We compare DSRC-QCS against five RAG approaches across three datasets and three LLMs. DSRC-QCS demonstrates superior performance. Compared to the Direct Retrieval method, DSRC-QCS improves the average F1 score by 17.36% with Alpaca-7b, 10.83% with LLaMa2-Chat-7b, and 11.88% with GPT-3.5-Turbo. We also conduct ablation studies to validate the performance of both DSRC and QCS and explore factors influencing the effectiveness of DSRC. We have included all prompts in the Appendix.

As the widespread use of algorithms and artificial intelligence (AI) technologies, understanding the interaction process of human–algorithm interaction becomes increasingly crucial. From the human perspective, algorithmic awareness is recognized as a significant factor influencing how users evaluate algorithms and engage with them. In this study, a formative study identified four dimensions of algorithmic awareness: conceptions awareness (AC), data awareness (AD), functions awareness (AF), and risks awareness (AR). Subsequently, we implemented a heuristic intervention and collected data on users' algorithmic awareness and FAT (fairness, accountability, and transparency) evaluation in both pre-test and post-test stages (N = 622). We verified the dynamics of algorithmic awareness and FAT evaluation through fuzzy clustering and identified three patterns of FAT evaluation changes: “Stable high rating pattern,” “Variable medium rating pattern,” and “Unstable low rating pattern.” Using the clustering results and FAT evaluation scores, we trained classification models to predict different dimensions of algorithmic awareness by applying different machine learning techniques, namely Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), and XGBoost (XGB). Comparatively, experimental results show that the SVM algorithm accomplishes the task of predicting the four dimensions of algorithmic awareness with better results and interpretability. Its F1 scores are 0.6377, 0.6780, 0.6747, and 0.75. These findings hold great potential for informing human-centered algorithmic practices and HCI design.

The Latin-American scientific community has achieved significant progress towards gender parity, with nearly equal representation of women and men scientists. Nevertheless, women continue to be underrepresented in scholarly communication. Throughout the 20th century, Latin America established its academic circuit, focusing on research topics of regional significance. Through an analysis of scientific publications, this article explores the relationship between gender inequalities in science and the integration of Latin-American researchers into the regional and global academic circuits between 1993 and 2022. We find that women are more likely to engage in the regional circuit, while men are more active within the global circuit. This trend is attributed to a thematic alignment between women's research interests and issues specific to Latin America. Furthermore, our results reveal that the mechanisms contributing to gender differences in symbolic capital accumulation vary between circuits. Women's work achieves equal or greater recognition compared to men's within the regional circuit, but generally garners less attention in the global circuit. Our findings suggest that policies aimed at strengthening the regional academic circuit would encourage scientists to address locally relevant topics while simultaneously fostering gender equality in science.