Information Processing and Management: an International Journal

• We assess how GitHub repositories influence citation rates of ML research papers.
• ML papers enjoy a 20% monthly citation boost after the first GitHub repositories.
• Frameworks popularity (e.g., PyTorch) also positively shapes ...

Open Science initiatives prompt machine learning (ML) researchers and experts to share source codes - "scientific artifacts" - alongside research papers via public repositories such as GitHub. Here we analyze the extent to which 1) the ...

• As Cognitive Overload increases, so does the Anxiety and the Avoidance Behavior.
• The effect of Cognitive Overload on Cognitive Fatigue is fully mediated by Anxiety.
• The effect of Anxiety on Avoidance Behavior is fully mediated by ...

We aim to investigate how Cognitive Overload, Anxiety, Cognitive Fatigue, Avoidance Behavior and Data Literacy are related in Big Data environments. We developed a survey with 372 respondents and analyze the data using Partial Least Squares ...

Display Omitted

In recent years, Zero-shot Node Classification (ZNC), an emerging and more difficult task is starting to attract attention, where the classes of testing nodes are unobserved in the training stage. Existing studies for ZNC mainly utilize Graph ...

• Robust multi-view enhanced node representations for enriching node information.
• Self-supervised contrastive learning method to create discriminative and unique representations of nodes.
• Effective knowledge-aware augmentation for ...

• 35 worldwide Open Government Data sites contained 1514 COVID-19 datasets.
• Datasets in 41 data formats, produced by organizations at six government levels.
• Correlations found between the number of data formats and datasets, and data ...

In this study, we investigate the key characteristics of more than 1500 COVID-19 datasets appearing on 35 Open Government Data (OGD) sites around the world. In addition to examining the number of datasets, data formats, star levels (a measure of ...

• Identifying causal effects in auto-encoding of free-text nursing notes.
• Investigating the impact of the clinical sentiment of patients' nursing notes on fall risk stratification.
• The dataset of the quasi-experiment consists of 2,...

Stratifying patients with high risks of falling based on assessment with nursing notes is essential for tailoring anticipated fall prevention strategies. However, the average exposure effects of the clinical sentiment polarity (CSP) of nursing ...

Stock trend prediction is crucial for recommending high-investment value stocks and can strongly assist investors in making decisions. In recent years, the significance of stock relationships has been gradually recognized for trend prediction, ...

• A dynamic graph construction module is proposed.
• A new distance algorithm for stock based on motif is proposed.
• Prior relationships are introduced to guide the dynamic graph construction.

Text, 2D, and 3D information are crucial information representations in modern science and management disciplines. However, complex and irregular 3D data produce data scarcity and expensive generation that limit their processing and application. ...

• The first 3D mesh processing approach by leveraging across-modal information in a zero/few-shot manner.
• A novel cross-modal learning strategy combining the modality vision data and 3D graphics data.
• A novel self-attention adapter ...

• Highlights the difference between UGC and MGC from an economic perspective.
• Empirically shows the differential influences of the textual features, communication modes and legitimacy of UGC and MGC on adverse events.
• Provides ...

The rapid dissemination of information on social media may bring potential risks to companies. However, the current discussions about the impact of social media are inadequate. This study aims to examine the moderating effect of social media ...

The cross-domain recommendation aims to make more accurate personalized recommendations by transferring user preferences from a relatively rich interactive information domain to a sparse data domain. Traditional methods mostly only use direct ...

• Balanced fuse user’s indirect side-information and effectively suppress the noise.
• Propose a user’s feature disentangle module to extract the domain-invariant features.
• Effectively enhance the rating prediction accuracy and ...

• We propose a group convolutional neural network for speech emotion recognition.
• We design an interpretability constraint for interpretable representation.
• Complementary autonomous representation is learned with an uncorrelation ...

This paper focuses on the active interpretability for deep learning-based speech emotion recognition (SER). To achieve this, we propose an explicit feature constrained model, the interpretable group convolutional neural network (IG-CNN) model. In ...

• A metaverse platform pricing model is developed to analyze the business model of Real-to-Virtual platforms.
• The type and strength of network effects determine the type of bundling strategies.
• The differences in bundling can reflect ...

This paper aims to investigate the motivations behind the reconstruction of the business model for Real-to-Virtual platforms in the metaverse era. By developing a pricing model for metaverse platforms that consider both social network effects and ...

The use of medical blockchain enables the safe sharing of electronic medical data (EMD). With the continuous growth of EMD, how to efficiently find EMD on the medical blockchain is a very challenging problem. In this paper, we propose an ...

• Our method supports ciphertext query services with different conditions.
• We design an access control scheme and a ciphertext search scheme to preserve privacy.
• We theoretically prove the security of our method and verify it on real ...

• Integrates AI teammates to a human-AI collaboration platform in a work team.
• Explore three capability dimensions: individual, organization, and cross-organization.
• Analyze a successful AI integration case study in a large ...

Artificial Intelligence (AI) can enter organizations and become AI teammates in organizational teams, posing new challenges to organizational form, team management, and team working patterns. Establishing an efficient "human-AI team" requires an ...

• Lead entrepreneur's digital leadership have a positive influence on the entrepreneurial team's digital entrepreneurial success.
• Entrepreneurial team's technology absorptive capacity and technological innovation capability mediate the ...

Digital entrepreneurial success has gained attention because the development of digitalization has brought multiple opportunities for entrepreneurs along with many challenges. While prior research has primarily examined the determinants of ...

Sentiment prediction is useful for scientific decision-making and reliable assessments in various fields. One significant challenge of sentiment prediction is the difficulty in dynamically capturing changing dependency patterns along the ...

• A Time-Enhanced Neighbor-Aware network is proposed for sentiment prediction in social networks.
• We introduce a time-enhanced LSTM model for handling irregular sequential data.
• The neighbor-aware encoder can obtain neighbors’ ...

• 26,965 articles in LIS journals were analyzed to identify research topics and categorize research methods.
• There is a shift in the research strategy from conceptual to empirical research in LIS investigations over the past 3 decades.

The present study analyzed over 26,000 research articles published between 1991 and 2021 in twenty-one major LIS (Library and Information Science) journals, using the machine learning (ML) approach to categorize the research methods used by LIS ...

Privileged information, a form of prior knowledge, can significantly enhance traditional machine learning performance through a novel paradigm known as learning using privileged information (LUPI). Although effective, current studies on LUPI ...

• Propose a brand new problem of learning using class-wise privileged information.
• Build a model called CGSVM+ by using the augmented class-wise privileged information.
• Collect two real-world datasets and annotate them by ...

Currently, the expressive power of momentum-driven graph Siamese representation learning is mainly constrained by flawed positive selection strategies and data augmentation strategies. To solve these issues, we propose a Semantics-preserved Graph ...

• In graph Siamese representation learning, disregarding the local structure during positive selection or applying augmentation directly to the original graph can result in erroneous positives and misguide the training process of the model.

Multi-modal Aspect-based Sentiment Analysis (MABSA) aims to forecast the polarity of sentiment concerning aspects within a given sentence based on the correlation between the sentence and its accompanying image. Comprehending multi-modal ...

• CoolNet enables an efficient cross-modal alignment and fusion.
• We are the first to harness both semantic and syntactic features for this task.
• We treat the image as a graph structure to model graph-level features.
• Our model ...

Table filling based joint named entity recognition and relation extraction task aims to share representation of subtasks in a table to extract structured knowledge. However, most of existing studies need additional labels and dedicated deep ...

• Propose a novel DREAM model for joint NER and RE task.
• Learn shared representations by a novel LSRL module.
• Extract purified task-specific features by a new TIB module.
• Performances of DREAM achieve the state-of-arts.

• A double-layer aspect recognition model is developed to recognize objects and aspects in online medical reviews, outperforming baseline approaches by 23.12%.
• We incorporate the expert knowledge of OMR-ontology into the proposed aspect-...

Online medical services have become increasingly popular, and patient feedback can significantly influence other patients’ medical decision-making. This study utilizes a double-layer domain ontology for conducting aspect-level sentiment analysis ...

Large language models (LLMs) are believed to contain vast knowledge. Many works have extended LLMs to multimodal models and applied them to various multimodal downstream tasks with a unified model structure using prompt. Appropriate prompts can ...

Event detection (ED) is aimed to identify the key trigger words in unstructured text and predict the event types accordingly. Traditional ED models are too data-hungry to accommodate real applications with scarce labeled data. Besides, typical ED ...

• We apply the true few-shot event detection to accommodate low-resource scenarios.
• We develop a multi-step prompt to focus on the event context.
• We introduce a prototypical network to mitigate the disabled generalization issue.
• ...

The potential impact of a paper is often quantified by how many citations it will receive. However, most commonly used models may underestimate the influence of newly published papers over time, and fail to encapsulate this dynamics of citation ...

• We construct hierarchical and heterogeneous graphs to record dynamics of the citation network .
• We propose a novel model H2CGL to aggregate structural and temporal features.
• The sensitivity of graph representations to potential ...

Session-based recommendation (SBR) originates from a real-world need to provide effective recommendation solutions for unlogged users. How to utilize short interaction sequences of anonymous users for practical recommendations has become a ...

• We propose a novel recommendation model based on GNN and sparse attention.
• Target information is learned by session target module and selectively correcting and enhancing it.
• An intent-collaboration module is proposed to obtain ...

• An event extraction corpus with document-level causality analysis oriented social security is constructed from 2235 web texts of Chinese news and micro-blog Weibo.
• A dynamic tagging process for event extraction corpus construction is ...

Event corpora are imperative to train event extraction models. Currently, most existing event corpora suffer from being available only in English, and their construction is limited by high annotation costs. This paper aims to construct a corpus ...

IP geolocation is necessary for applications such as location-aware ad recommendation, traceability, and fraud detection. Presently, graph neural network-based approaches focus on geolocation information transfer of neighboring nodes of the ...

Electronic sports (eSports) is competitive video gaming that is coordinated and managed by sporting organizations. While traditional sports have thrived on spectatorship and the intense emotional experiences of fans, there has been limited ...

• A mixed-methods methodology for investigating eSports spectatorship is presented.
• Social interactions and emotional dimensions are investigated through network-based and thematic analyses.
• Social media discussions of a real eSport ...

Automatic sarcasm detection from text is one important research task in text mining and natural language processing and has attracted extensive attention from researchers. Most approaches focus on designing various models and features according ...