Collaborative Information Research Articles

Collaborative Sequential Recommendations via Multi-view GNN-transformers

Sequential recommendation systems aim to exploit users’ sequential behavior patterns to capture their interaction intentions and improve recommendation accuracy. Existing sequential recommendation methods mainly focus on modeling the items’ chronological relationships in each individual user behavior sequence, which may not be effective in making accurate and robust recommendations. On the one hand, the performance of existing sequential recommendation methods is usually sensitive to the length of a user’s behavior sequence (i.e., the list of a user’s historically interacted items). On the other hand, besides the context information in each individual user behavior sequence, the collaborative information among different users’ behavior sequences is also crucial to make accurate recommendations. However, this kind of information is usually ignored by existing sequential recommendation methods. In this work, we propose a new sequential recommendation framework, which encodes the context information in each individual user behavior sequence as well as the collaborative information among the behavior sequences of different users, through building a local dependency graph for each item. We conduct extensive experiments to compare the proposed model with state-of-the-art sequential recommendation methods on five benchmark datasets. The experimental results demonstrate that the proposed model is able to achieve better recommendation performance than existing methods, by incorporating collaborative information.

Graph-Enhanced Prompt Learning for Personalized Review Generation

Personalized review generation is significant for e-commerce applications, such as providing explainable recommendation and assisting the composition of reviews. With the success of pre-trained language models (PLMs), prompt learning-based approaches have been employed to handle this task. However, the existing approach neglects the historical user-item interactions as well as the diverse semantics of the reviews (including semantically relevant reviews and semantically irrelevant reviews). In this paper, we propose GRAPA, a graph-enhanced prompt learning approach for personalized review generation. Specifically, GRAPA extracts topic-level information for each review to address the semantic diversity of reviews. Moreover, GRAPA employs a heterogeneous graph neural network (GNN) to explore the collaborative information hidden in historical user-item interactions. User and item representations generated by the GNN module as well as their ID embeddings are used as prompts and fed into a PLM to guide the generation process. To alleviate the interference of semantically irrelevant reviews, GRAPA further proposes a contrastive learning module to distinguish them. Experimental results on public datasets show that GRAPA outperforms existing methods by up to 4.3% in BLEU-4 and 5.4% in ROUGE2-F.

The role of the innovation ecosystem as a tool for innovation development

The development of various industrial sectors is one ofthe priority tasks of the state, the solution of which is toimprove the quality and life expectancy of the population.According to the authors of the article, an innovationecosystem should be understood as an open dynamicsystem that provides collaborative forms of interaction between actors, taking into account the non-hierarchical relationships that have developed within the institutional ecosystem of enterprises, regions, and countries acting inthe direction of sustainable advanced innovativedevelopment based on modern technological concepts.Modern innovation ecosystems are created and functionto overcome individual limitations, maximise resourceuse, and distribute risks and costs to better achievecommon goals. This approach differs from the traditionalones in that it takes into account the current level ofeconomic development, taking into account the processesof digital transformation of all spheres of activity andprovides an integrated approach to the formation of aninnovation ecosystem. The article concludes that industry is currentlydeveloping as a traditional innovation system. The analysis of different views and main approaches tounderstanding innovation ecosystems allowed to identifythe advantages of ecosystems in comparison withtraditional innovation systems and to justify the need todevelop industry as an innovation ecosystem. A moredetailed study was carried out on the basis of market datafor various types of equipment. These conclusions wereconfirmed when analysing the prerequisites for theformation of an innovation ecosystem based on theresults of the analysis of the current state of theequipment market. A number of systemic problems canbe identified that are common to all stages of productionand circulation of various products: limited financialresources of manufacturers for development (researchand development, sales and promotion of products);scientific and technological lagging behind the level ofworld leaders; difficulty in commercialisingdevelopments. These problems have a complex impact on thedevelopment of the industry as a whole and its individual product segments. Eliminating these constraints will helpcreate favourable conditions for the machinery marketand unlock its potential.

Neither Mergers nor Cartels: Innovation Networks and Competition Law

Abstract Organization theorists and innovation economists have been arguing for the past few decades that the locus of research and development activities has gradually moved away from vertically integrated firms to inter-firm alliances. In industries as diverse as construction and biotechnology, cooperation for innovation has given rise to collaborative forms of industrial organization varying from bilateral alliances to networks. Nonetheless, the EU competition law framework opens up limited space for inter-firm collaborations, cabined by the efficiency test of Article 101(3) TFEU and the requirements of the horizontal block exemption regulations. Building on this misfit, this article makes two contributions. First, a theoretical one, arguing that the law’s cautious approach to inter-firm collaboration is rooted in a binary vision of industrial organization which recognizes only firms and markets as modes of organizing production. Second, an explanatory one, showing how the 2023 Research & Development Block Exemption Regulation falls short of facilitating inter-firm collaborations for innovation and proposing fine-tuning suggestions.

From Boardrooms to Classrooms: How Interorganizational Networks Influence Education Policy Adoption

This paper examines how relationships among government and “outside” organizations influenced policy implementation of new dropout prediction data systems. Using comparative historical and network analyses of three cities, I suggest the concept of interorganizational coupling—highlighting how the dependence and (in)formal collaborations among local school improvement organizations affected implementation speed, variation, and constraint. In Chicago, the loosely coupled system influenced slow and varied implementation, sustained by interpersonal relations and challenged by unclear division of labor. In Philadelphia, the tightly coupled system shaped swift and uniform changes, constrained by questions of sustainability. In New York, the tightening system led to fast yet variable transformations, limited by competition among organizations. Broadly, the article contributes to studies of education policies, interorganizational networks, and school improvement.

A novel similarity measure SF-IPF for CBKNN with implicit feedback data

PurposeThe increasing popularity of music streaming services also increases the need to customize the services for each user to attract and retain customers. Most of the music streaming services will not have explicit ratings for songs; they will have only implicit feedback data, i.e user listening history. For efficient music recommendation, the preferences of the users have to be infered, which is a challenging task.Design/methodology/approachPreferences of the users can be identified from the users' listening history. In this paper, a hybrid music recommendation system is proposed that infers features from user's implicit feedback and uses the hybrid of content-based and collaborative filtering method to recommend songs. A Content Boosted K-Nearest Neighbours (CBKNN) filtering technique was proposed, which used the users' listening history, popularity of songs, song features, and songs of similar interested users for recommending songs. The song features are taken as content features. Song Frequency–Inverse Popularity Frequency (SF-IPF) metric is proposed to find the similarity among the neighbours in collaborative filtering. Million Song Dataset and Echo Nest Taste Profile Subset are used as data sets.FindingsThe proposed CBKNN technique with SF-IPF similarity measure to identify similar interest neighbours performs better than other machine learning techniques like linear regression, decision trees, random forest, support vector machines, XGboost and Adaboost. The performance of proposed SF-IPF was tested with other similarity metrics like Pearson and Cosine similarity measures, in which SF-IPF results in better performance.Originality/valueThis method was devised to infer the user preferences from the implicit feedback data and it is converted as rating preferences. The importance of adding content features with collaborative information is analysed in hybrid filtering. A new similarity metric SF-IPF is formulated to identify the similarity between the users in collaborative filtering.

Highlanders: a qualitative study to explore the experiences of home care nurses in rural areas in Italy

Introduction and aim: Home care in rural areas is characterised by peculiarities linked to both the type of care of the patient and his/her family and the context in which it is provided. International studies have highlighted the need for specialised training and advanced clinical skills to work in this field. Currently, there are no studies in the literature that explore the experiential feelings of Italian nurses who provide home care in rural areas. Setting: Rural areas in Italian Apennine District. Participants: Italian home care nurses in the selected areas. Design: Qualitative study using a transcendental phenomenological approach. Semi-structured interviews were conducted face-to-face, and audio was recorded with home care nurses. Results: Eleven nurses (mean age, 53 years) were interviewed. Four themes supported by nine codes emerged: "Entering your homes" (Stepping lightly; a matter of trust; key professional); "Being home care nurses in rural areas" (Working alone, autonomy, responsibility and gratifications; ideal characteristics of home care nurse); "Love and hate toward doctors" (Family doctors; ANT doctors); "The territory: the Apennines and mountainous areas" (working in rural areas; healthcare professionals).Discussion and conclusions: The risks for home careers in rural areas are related to the vastness of the territory, long car journeys, the harsh climate in winter, and pets in the homes. The beauty of the area, the close relationship established with the families in charge, and daytime working hours are reasons for great professional and personal satisfaction. Some modalities and forms of collaboration with general practitioners can be improved.

Graph Neural Networks for Recommendation Leveraging Multimodal Information

Recommender systems act as filtering algorithms to provide users with items that might meet their interests according to the expressed preferences and items' characteristics. As of today, the collaborative filtering paradigm, along with deep learning techniques to learn high-quality users' and items' representations, constitute the de facto standard for personalized recommendation, showing remarkable recommendation accuracy performance. Nevertheless, recommendation remains a highly-challenging task. Among the most debated open issues in the community, this thesis considers two algorithmic and conceptual ones, namely: (i) the inexplicable nature of users' preferences, especially when they come in the form of implicit feedback; (ii) the effective exploitation of the collaborative information in the designing and training of recommendation models. In domains such as fashion, food, and media content recommendation, the shallow item's profile can be enhanced through the multimodal characteristics describing items [Malitesta et al., 2023]. Driven by these assumptions, in the first part of this thesis, we apply multimodal deep learning strategies for multimedia recommendation; the scope is to study and design recommendation algorithms based upon the principles of multimodality to possibly match each item's characteristic to the implicit preference expressed by the user [Deldjoo et al., 2022], thus addressing the (i) issue. Recent collaborative filtering approaches profile users and items through embedding vectors in the latent space. However, such models disregard structural properties naturally encoded into the user-item interaction data. Indeed, recommendation datasets are easily describable under the topology of a bipartite and undirected graph, with users and items being the graph nodes connected at multiple distance hops. In this respect, the application of graph neural networks , recent deep learning techniques specifically tailored to learn from non-euclidean data, can provide a refined representation of users and items to mine near- and long-distance relationships in the user-item graphs [Anelli et al., 2023b]. Indeed, this is one possible solution to exploit the collaborative information, which is effectively propagated within the user-item graph, addressing the (ii) issue. Conclusively, this thesis aims to match the two families of recommendation strategies by leveraging graph neural networks and multimodal information data [Anelli et al., 2022]. In doing so, other numerous micro-aspects within the two macro-areas (introduced above) are examined. Indeed, the thesis is a systematic compendium of careful analyses regarding, among others, reproducibility, novel evaluation dimensions [Anelli et al., 2023a], and tasks/scenarios complementary to recommendation. Awarded by : Politecnico di Bari, Bari, Italy on 30 January 2024. Supervised by : Tommaso Di Noia. Available at : https://hdl.handle.net/11589/264941.

With instead of about: Toward an anthropology that is critically integrated into Global Early Childhood Development interventions

AbstractThis comment posits questions and opens debates around the recent article by Scheidecker et al. based in the author's experience in anthropological research and applied work in child development in the Global South. The article advocates for a critical review of how anthropologists in the Global South carry out and disseminate their research, in order to broaden audiences beyond the academy. Finally, it is argued that for anthropological evidence to hold a place at the table on Global Early Childhood Development (ECD), anthropologists must engage in practices, methodologies, and forms of collaboration that make our findings and perspectives hearable. For this to happen, anthropologists should take an active part in institutions and fields of work they tend to avoid. Only in this way can we positively impact children and families by incorporating the inherent diversity of ECD beyond the parameters and values of normalcy that predominate in the so‐called minority world.

ANALYSIS OF THE BIM IMPLEMENTATION COMPETENCIES OF BUILDING CLUSTER VOCATIONAL HIGH SCHOOL STUDENTS IN THE EAST JAVA

Even though the government, through the Ministry of Public Works and Public Housing (PUPR), introduced the concept of Building Information Modeling (BIM) in regulations as a form of collaboration and integration for application in the Indonesian construction industry in 2017, the adoption rate of BIM in Indonesia remains low. There are many obstacles that must be faced, such as the curriculum at the vocational high school level to higher education, or skill sets that have not yet been developed in their users. This study examines the abilities of Building Cluster Vocational High School Students in terms of mastered skills that are in line with BIM competency standards. The methods are a survey and the use of the Relative Importance Index on the participants of the 30th Vocational High School Competency Competition (LKS) at the East Java Provincial Level in 2022. The finding indicates that non-cognitive competencies dominate the adaptability of these students' mastery of BIM software.

Multi-perspective learning for enhanced user preferences for session-based recommendation

Due to the sparsity and anonymity of interactions within sessions, capturing user preferences by modeling item relationships poses a significant challenge. Previous methods have attempted to utilize collaborative information from adjacent sessions or item attribute knowledge as supplements. However, these methods are prone to introducing interfering information through their session similarity measurement strategies and attribute embedding learning processes, both requiring optimization. Moreover, they often fail to fully model global item relationships. To address these issues, we propose a novel method called Multi-perspective Learning for Enhanced User Preferences (MLEUP) for session-based recommendations. Specifically, we first construct reliable session groups by measuring the overlap and incompatibility of items across sessions to calculate session similarity. Then, by modeling attribute–item and attribute–attribute relationships, along with employing a neighbor-based contrastive self-supervised attribute learning task, we learn high-quality attribute embeddings that capture the user’s stable preferences. Next, leveraging the session groups, we model global item multi-granularity co-occurrence and high-order sequential relationships through designed attention layers, enabling us to learn informative item embeddings that represent the user’s dynamic preferences. Finally, we further model the correlated relationship between the two user preference representations. Our approach ensures that effective attribute knowledge and global collaborative information are learned to mitigate the data sparsity issue, while comprehensively modeling global item relationships to capture the user’s complex dynamic preferences. Experimental results demonstrate that MLEUP achieves state-of-the-art performance in terms of precision, MRR, and NDCG. The experimental datasets and source code of MLEUP are publicly available at https://github.com/lean030/MLEUP.

The challenge of adopting a collaborative information system for independent healthcare workers in France: a comprehensive study

This study aimed to explore the perception of an underutilised collaborative information system through qualitative research, utilizing semi-structured, in-depth interviews with independent midwives and physician. PROSPERO, is a collaborative information system designed to bridge the communication gap between community-based healthcare workers and hospital-based care teams for parturients in Lyon, France. Through 27 semi-structured in-depth interviews with midwives, obstetricians, and general practitioners, we identified key themes related to the system’s adoption: implementation challenges, utilisation barriers, interprofessional dynamics, and hidden variables affecting system use. Participants recognised the potential of PROSPERO to improve information sharing and care coordination but expressed concerns about the system’s integration into existing workflows, time constraints, and the need for adequate training and technical support. Interprofessional dynamics revealed differing perspectives between hospital and independent practitioners, emphasising the importance of trust-building and professional recognition. Hidden variables, such as hierarchical influences and confidentiality concerns, further complicated the system’s adoption. Despite the consensus on the benefits of a collaborative information system, its implementation was hindered by mistrust between healthcare workers (i.e. between independent practitioners and hospital staff). Our findings suggest that fostering trust and addressing the identified barriers are crucial steps towards successful system implementation. The study contributes to understanding the complex interplay of factors influencing the adoption of collaborative healthcare technologies and highlights the need for strategies that support effective interprofessional collaboration and communication.ClinicalTrials ID NCT02593292.

Conceptualising collaboration for educational change: the role of leadership and governance

ABSTRACT This paper critically examines the concept of collaboration, and how it can be mobilised in policy and practice in the pursuit of educational change in the context of the Scottish education system. Policy analysis and interviews with primary school headteachers highlighted the lack of definitional agreement but common conceptual characteristics of collaboration, and the implications of this for how it is led and governed in the pursuit of change. Highlighted in this paper is the complex interplay between forms of, influences on, and drivers of collaboration and how this results in varied outcomes of it. It is argued that collaboration across systems should be focused, purposeful, and supported with high levels of trust from those involved or enabling it to happen, and meso- and macro-level structures which are flexible to enable diverse forms of collaboration to emerge. This paper presents a framework for collaboration that supports the complex consideration of the interrelated domains of forms, influences, and drivers of collaboration, and the leadership practices and governance arrangements necessary for collaboration to lead to improvement and change.

KGCFRec: Improving Collaborative Filtering Recommendation with Knowledge Graph

Traditional collaborative filtering (CF)-based recommendation systems are often challenged by data sparsity. The recent research has recognized the potential of integrating new information sources, such as knowledge graphs, to address this issue. However, a common drawback is the neglect of the interplay between user–item interaction data and knowledge graph information, resulting in insufficient model performance due to coarse-grained feature fusion. To bridge this gap, in this paper, we propose a novel graph neural network (GNN) model called KGCFRec, which leverages both Knowledge Graph and user–item Collaborative Filtering information for an enhanced Recommender system. KGCFRec employs a dual-channel information propagation and aggregation mechanism to generate distinct representations for the collaborative knowledge graph and the user–item interaction graph. This is followed by an attention mechanism that adaptively fuses the knowledge graph with collaborative information, thereby refining the representations and narrowing the gap between them. The experiments conducted on three real-world datasets demonstrate that KGCFRec outperforms state-of-the-art methods. These promising results underscore the capability of KGCFRec to enhance recommendation accuracy by integrating knowledge graph information.

Optimizing Urban Mobility: Assessing the Impact of Selangor's Intelligent Transport System (SITs)

This paper addresses Malaysia's public transport system, primarily managed by small operators in its developing state. Efficient transportation management relies on collaborative information sharing. Systems must provide an easy and effective way to manage staff and registered users, including students and community members. Working passengers require efficient route information to minimize delays. Therefore, systems should cater to all users' needs. Registered users, such as students, also require real-time bus updates to avoid long waits and inform the organization of delays via SMS. Implementing and managing these services collectively presents challenges. Mobile technology, including smartphones and laptops, offers a solution through location-based applications for managing transportation with map navigation. The recent mobile technology revolution has intensified competition among developers, resulting in similar yet distinct applications. Modern transit services now offer virtual access to schedules and advanced features like integration with Google Maps and real-time tracking. De-spite benefits, implementing a Smart Transportation Application (STA) with GPS and digital route maps can be daunting and costly for smaller operators.

Learning to teach at a norm-critical clinical learning centre: A Phenomenographic study

BackgroundCampus-based clinical learning centres are used for practice and learning in nursing students' education and can be arenas in which to enhance students' awareness and competence in social justice issues. Norm-critical approaches can be used as pedagogical tools in these centres to prepare students for hands-on caring situations in which social norms can bias the outcome. ObjectivesTo describe nursing teachers' conceptions of learning norm-critical approaches and implementing them in a clinical training centre. Setting and participantsThe study is based on interviews with 10 teachers at a Swedish university college. MethodsThe data was analysed using a phenomenographic approach. ResultsFive categories of description emerged in the analysis that described conceptions related to norm-critical approaches. These categories were: personally developing and meaningful; easily integrated with established nursing concepts; highlighting surrounding power; something to lean on when letting students take the first steps in norm-critical initiatives; and helping implementation in teaching and education. ConclusionsTeachers who are facing the task of providing norm-critical, practical education to nursing students in campus-based clinical learning environments are ambivalent towards the core principles of norm criticism, which they conceive as natural and provoking, at the same time. We therefore need to un-dramatise norm criticism and better prepare teachers in how to use it. Teachers would benefit from follow-up activities and formal collaborations regarding norm-critical teaching, so that they are given context to discuss, reflect, and learn from each other.

Sociotechnical Imaginaries of Sharing and Emerging Postdigital Meaning-Making Practices in the Astronomy Community

For decades, science communities have had digital technology embedded in their everyday work. However, new research infrastructures are amplifying the presence and use of digital technologies for scientists. In that respect, radio astronomy is undergoing a major transformation causing the community to enter a phase of postdigital work, due to the construction of the most sensitive telescope, the Square Kilometer Array Observatory (SKAO), which scales the presence of digital technology and the amount of data generated. As new digital research infrastructure is set up, sociotechnical imaginaries—symbols and visions of a shared future—emerge, while others become obsolete thus impacting structures and practices of meaning-making. In this paper, we explore the disruptive potential of sociotechnical imaginaries and how astronomers using SKA pathfinder and precursor telescope data respond to these imaginaries and incorporate them into their meaning-making. The analysis shows that postdigital imaginaries related to data circulation, storage, archiving, and reuse have been amplified as SKA facilities and services are set up. Two changes are highlighted regarding a new postdigital condition within the astronomy community. Firstly, as astronomers engage in new postdigital forms of collaboration, they need to reach a consensus on what types of analyses to use by agreeing which methods are appropriate. This affects how scientific questions and research proposals are negotiated collectively, impacting the agency of astronomers. Secondly, as digital tools are increasingly part of astronomers’ daily work, they have to rely on new data analysis methods, which determine what evidence is uninteresting. Overall, these changes pose new questions regarding how meaning-making processes are altered and the way science is undertaken because of these new entanglements of human and non-human actors.

“Oh Yes! Net-Zero”: Sociotechnical capabilities and regional innovation systems for British industrial decarbonization

Few sectors in the global economy need deep decarbonization as much as heavy industry, which is currently the largest and fastest growing source of global carbon emissions. Based on an original dataset of 111 expert interviews and 52 site visits, this paper asks: what sociotechnical pathways, capabilities, and regional innovation systems are emerging to support industrial decarbonization? It combines and applies insights from evolutionary economic geography and transitions studies, such as sociotechnical capabilities and learning, as well as spatial frameworks such as challenge-oriented regional innovation systems (CoRIS) and green industrial restructuring. It utilizes these collective concepts of pathways, capabilities and CoRIS to explore the organizational and technological, individual, and systems level dimensions of industrial decarbonization efforts ongoing across Great Britain. We find that within industrial net-zero clusters, spatial and technological proximities are being effectively leveraged, especially through new forms of networking, collaboration, and partnerships. CoRIS provides the basis for greater reflection on perennial transitions issues, both within and across clusters, related to incumbency and innovation, in addition to spotlighting emergent issues such as light versus deep green restructuring and incremental versus transformative change. In simpler terms, industrial decarbonization unfolds as a multi-scalar process. Industrial decarbonization exhibits both incremental and transformative patterns of change. Industrial decarbonization lastly involves dynamic processes that coevolve across a spectrum of sociotechnical attributes.

Cross-domain contrastive graph neural network for lncRNA–protein interaction prediction

Identifying lncRNA–protein interactions (LPIs) is an important biomedical task, facilitating the comprehension of the biological functions and mechanisms of lncRNAs. Many computational methods have been developed for this task, especially graph neural network (GNN)-based methods have attracted increasing attention. Typically, the LPI network involves two types of interaction domains: the interactive domain capturing the direct interaction information between lncRNAs and proteins, and the collaborative domain reflecting the collaboration information among lncRNAs or proteins. However, existing GNN-based methods only leverage one of them to obtain topological information, which cannot fully characterize lncRNAs and proteins, resulting in suboptimal node representations. Moreover, each domain contains task-irrelevant redundant information, posing a challenge in effectively integrating information from different domains. To address these issues, we propose a novel Cross-domain Contrastive Graph Neural Network (CCGNN) for predicting potential LPIs. CCGNN employs a multi-domain encoder that consists of an interactive domain encoder and two collaborative domain encoders to capture valuable information from each interaction domain. Subsequently, domain-adaptive fusion is designed to integrate information from different domains to acquire comprehensive node representations. Furthermore, cross-domain contrastive learning is devised to enrich the node representations, drawing inspiration from the information bottleneck principle by retaining as much task-relevant information as possible within each domain and minimizing mutual information between representations across different domains. Extensive experiments on four real-world datasets demonstrate the superiority of CCGNN over state-of-the-art methods, and a further case study and generalization analysis illustrate the effectiveness of CCGNN in the biomedical link prediction tasks.

Cooperative Learning in an Interior Architecture Studio

Design studio in interior architecture education is rarely conducted as a group learning activity, despite the benefits of preparing students for team-based industry work environments. Students have learnt to be protective of their design ideas and are not always encouraged to share resources, although theories on social learning indicate a deeper learning experience is achieved through collaboration. The theoretical framework for cooperative learning, as a form of collaboration, is extensive and positively supports the virtues of group work as an effective learning activity. This paper describes an action research cycle of preparation, implementation and reflection of a cooperative learning project with third year interior architecture students. For the purposes of this research, cooperative learning is defined as a structured group work activity, with participants working together towards a series of common objectives. The main objectives were for students to experience a typical interior architecture work environment, to promote deeper learning and foster a culture of sharing ideas and resources.