REGAL

Network alignment (NA) offers a comprehensive way to build associations between different networks by identifying shared nodes. While the majority of current NA methods rely on the topological consistency assumption, which posits that shared nodes across different networks typically have similar local structures or neighbors, we argue that anchor nodes, which play a pivotal role in NA, face a more challenging scenario that is often overlooked. In this paper, we conduct extensive statistical analysis across networks to investigate the connection status of labeled anchor node pairs and categorize them into four situations. Based on our analysis, we propose an end-to-end network alignment framework that uses node representations as a distribution rather than a point vector to better handle the structural diversity of networks. To mitigate the influence of specific nodes, we introduce a mask mechanism during the representation learning process. In addition, we utilize meta-learning to generalize the learned information on labeled anchor node pairs to other node pairs. Finally, we perform comprehensive experiments on both real-world and synthetic datasets to confirm the efficacy of our proposed method. The experimental results demonstrate that the proposed model outperforms the state-of-the-art methods significantly.

Analogous to genomic sequence alignment, biological network alignment identifies conserved regions between networks of different species. Then, function can be transferred from well- to poorly-annotated species between aligned network regions. Network alignment typically encompasses two algorithmic components: node cost function (NCF), which measures similarities between nodes in different networks, and alignment strategy (AS), which uses these similarities to rapidly identify high-scoring alignments. Different methods use both different NCFs and different ASs. Thus, it is unclear whether the superiority of a method comes from its NCF, its AS, or both. We already showed on state-of-the-art methods, MI-GRAAL and IsoRankN, that combining NCF of one method and AS of another method can give a new superior method. Here, we evaluate MI-GRAAL against a newer approach, GHOST, by mixing-and-matching the methods' NCFs and ASs to potentially further improve alignment quality. While doing so, we approach important questions that have not been asked systematically thus far. First, we ask how much of the NCF information should come from protein sequence data compared to network topology data. Existing methods determine this parameter more-less arbitrarily, which could affect alignment quality. Second, when topological information is used in NCF, we ask how large the size of the neighborhoods of the compared nodes should be. Existing methods assume that the larger the neighborhood size, the better. Our findings are as follows. MI-GRAAL's NCF is superior to GHOST's NCF, while the performance of the methods' ASs is data-dependent. Thus, for data on which GHOST's AS is superior to MI-GRAAL's AS, the combination of MI-GRAAL's NCF and GHOST's AS represents a new superior method. Also, which amount of sequence information is used within NCF does not affect alignment quality, while the inclusion of topological information is crucial for producing good alignments. Finally, larger neighborhood sizes are preferred, but often, it is the second largest size that is superior. Using this size instead of the largest one would decrease computational complexity. Taken together, our results represent general recommendations for a fair evaluation of network alignment methods and in particular of two-stage NCF-AS approaches.

Cross-scale contrastive triplet networks for graph representation learning

Learning node representations against perturbations

Network alignment is an important technique with applications in diverse domains, such as social network analysis, bioinformatics, and knowledge graph construction. Many of the alignment methods rely on predefined anchor nodes, which are often unavailable in real-world scenarios. To overcome this limitation, we propose MANNA (MAtched Neighbor consistency for Network Alignment), an unsupervised approach to network alignment that exploits the concept of Matched Neighborhood Consistency (MNC). The hypothesis of MANNA is that nodes with higher similarity within their local neighborhood structures are more likely to be aligned across different networks. To learn the structural and attribute features of networks, MANNA uses a Graph Neural Network (GNN). It extracts multi-order node embeddings to capture multi-scale neighborhood features, which are then used to construct similarity matrices for the alignment process. MANNA introduces a key innovation by using pseudo-anchor nodes identified by the MNC strategy to provide self-supervised learning signals in the absence of real anchor nodes. This approach enhances the model’s ability to learn accurate network representations and improve alignment accuracy. Alignment results are iteratively refined by applying the MNC strategy, which strengthens the consistency of neighborhood structures between matched nodes. Extensive experiments on three public datasets show that MANNA outperforms existing network alignment methods.

DynHEN: A heterogeneous network model for dynamic bipartite graph representation learning

Graph representation learning has become a mainstream method for fraud detection due to its strong expressive power, which focuses on enhancing node representations through improved neighborhood knowledge capture. However, the focus on local interactions leads to imbalanced transmission of global topological information and increased risk of node-specific information being overwhelmed during aggregation due to the imbalance between fraud and benign nodes. In this paper, we first summarize the impact of topology and class imbalance on downstream tasks in GNN-based fraud detection, as the problem of imbalanced supervisory messages is caused by fraudsters' topological behavior obfuscation and identity feature concealment. Based on statistical validation, we propose a novel dual-view graph representation learning method to mitigate Message imbalance in Fraud Detection (MimbFD). Specifically, we design a topological message reachability module for high-quality node representation learning to penetrate fraudsters' camouflage and alleviate insufficient propagation. Then, we introduce a local confounding debiasing module to adjust node representations, enhancing the stable association between node representations and labels to balance the influence of different classes. Finally, we conducted experiments on three public fraud datasets, and the results demonstrate that MimbFD exhibits outstanding performance in fraud detection.

In this paper, we survey algorithms that perform global alignment of networks or graphs. Global network alignment aligns two or more given networks to find the best mapping from nodes in one network to nodes in other networks. Since graphs are a common method of data representation, graph alignment has become important with many significant applications. Protein-protein interactions can be modeled as networks and aligning these networks of protein interactions has many applications in biological research. In this survey, we review algorithms for global pairwise alignment highlighting various proposed approaches, and classify them based on their methodology. Evaluation metrics that are used to measure the quality of the resulting alignments are also surveyed. We discuss and present a comparison between selected aligners on the same datasets and evaluate using the same evaluation metrics. Finally, a quick overview of the most popular databases of protein interaction networks is presented focusing on datasets that have been used recently.

Network alignment finds node correspondences across multiple networks, where the alignment accuracy is of crucial importance because of its profound impact on downstream applications. The vast majority of existing works focus on how to best utilize the topology and attribute information of the input networks as well as the anchor links when available. Nonetheless, it has not been well studied on how to boost the alignment performance through actively obtaining high-quality and informative anchor links, with a few exceptions. The sparse literature on active network alignment introduces the human in the loop to label some seed node correspondence (i.e., anchor links), which are informative from the perspective of querying the most uncertain node given few potential matchings. However, the direct influence of the intrinsic network attribute information on the alignment results has largely remained unknown. In this paper, we tackle this challenge and propose an active network alignment method (Attent) to identify the best nodes to query. The key idea of the proposed method is to leverage effective and efficient influence functions defined over the alignment solution to evaluate the goodness of the candidate nodes for query. Our proposed query strategy bears three distinct advantages, including (1) effectiveness, being able to accurately quantify the influence of the candidate nodes on the alignment results; (2) efficiency, scaling linearly with 15 − 17 × speed-up over the straight-forward implementation without any quality loss; (3) generality, consistently improving alignment performance of a variety of network alignment algorithms.

Introduction The term “scientist” lacks a universally accepted definition, reflecting the evolving, interdisciplinary nature of scientific work and posing challenges for recognition, communication, and policy. This study aims to develop consensus-based definitions of the term “scientist” by engaging experienced scholars across diverse fields. Methods This study involved 156 scholars, each with at least 1,000 citations, recruited via convenience sampling. Fourteen scientist definitions, derived from literature and expert input, were assessed using a nine-point Likert scale via a structured google forms survey. The sample size was calculated using G*power (effect size = 0.5, power = 0.95), requiring at least 80 participants. Content Validity Index (CVI) was used for analysis. Definitions scoring ≥0.78 were accepted and included for final analysis, 0.70–0.78 were revised and re-evaluated, and <0.70 were excluded. Participation was voluntary and anonymous, ensuring ethical compliance and confidentiality. Results Of the 14 proposed definitions, six (42.9%) were excluded (CVI < 0.70), seven (50.0%) were accepted (CVI > 0.78), and one (7.1%) underwent revision (CVI 0.70–0.78). The highest-rated definitions were refined into two consensus-based versions: a short definition (“A scientist is a person who conducts research”) and a detailed one emphasizing hypothesis formulation and knowledge dissemination. Final validation yielded CVIs of 0.82 and 0.84, respectively, confirming strong expert agreement on both definitions. Conclusion This study developed two validated definitions of “scientist” emphasizing systematic research and knowledge dissemination. These definitions clarify the concept of scientific identity, providing a flexible yet rigorous framework applicable across academic, interdisciplinary, and policy-making contexts. Introduction The term "scientist" has undergone significant transformation since its inception, reflecting the dynamic nature of scientific inquiry and the evolving landscape of knowledge. This lack of clarity stems from the diverse roles and contributions of individuals in scientific fields, the evolving nature of research, and the interdisciplinary scope of modern science. Historically, figures such as Galileo and Newton were regarded as natural philosophers, a reflection of an earlier framework for knowledge production that has evolved alongside modern scientific advancements. Before twentieth century, the term "scientist" was commonly referred to as a "man of science," "natural philosopher," or by various other designations [1,2]. In contemporary contexts, scientists operate across a broad spectrum of fields, including medicine, biology, chemistry, physics, and social sciences, each employing methodologies tailored to their specific inquiries. For instance, biologists may design experiments to test hypotheses about living organisms, while social scientists might use qualitative methods to explore human behavior [3]. The Science Council defines a scientist as an individual who methodically collects and applies research and evidence to develop hypotheses, performs experiments, and shares results to advance knowledge in their field [4]. While National Cancer Institute defines a scientist as an individual with a background in science, particularly someone actively engaged in a specific area of research [5]. This diversity in practices underscores the challenge of defining "scientist" in a way that captures the breadth of their contributions. The plurality of definitions extends to global organizations and frameworks. For example, the United Nations Educational, Scientific, and Cultural Organization highlights the critical role of scientists in addressing global challenges and promoting sustainable development. This definition broadens the scope to include individuals working in multidisciplinary teams or applying scientific knowledge to public policy and societal issues. Similarly, some academic discussions focus on the characteristics of a scientist, such as curiosity, skepticism, and a commitment to evidence-based conclusions, rather than formal qualifications or job titles [6]. Unlike well-defined professions such as medicine or engineering, where specific educational pathways and professional titles (e.g., "doctor" or "engineer") confer clear identities, the term "scientist" lacks a universally recognized credentialing system. This absence can lead to underrepresentation or misrepresentation of scientific expertise, especially in interdisciplinary and collaborative contexts [7]. For example, the growing integration of data science in biology or physics illustrates the importance of understanding who qualifies as a scientist to ensure effective communication and collaboration among stakeholders. The absence of a standardized definition poses practical challenges for scientific communication, policymaking, and inclusivity. This study aims to address this gap by engaging scholars across disciplines to develop a consensus-based definition of "scientist." By recognizing the diverse and interdisciplinary contributions of scientists, such a definition could enhance collaboration, improve public understanding, and inform policies that support the scientific community. Methods Study design and participants A total of 156 scholars (out of 300 invited) participated in this study. Eligibility was determined based on the scholars' substantial academic expertise, evidenced by the achievement of at least 1,000 citations within their respective fields. This criterion ensured that participants had significant research experience and were highly qualified to contribute to the formulation of a consensus-based definition of "scientist." Participants were recruited through a convenience sampling method, and data were collected via a structured survey administered through google forms. While convenience sampling was used due to the accessibility of high-citation scholars, efforts were made to ensure disciplinary diversity to mitigate potential bias. Personalized invitations were sent via email to each scholar to facilitate their inclusion in the study. Sample size determination The sample size was determined using G*power statistical software (version 3.1.9.7), employing a two-tailed goodness of fit test with an effect size of 0.5, an alpha error probability of 0.05, and a statistical power of 0.95. According to the calculations, a minimum of 80 participants were required to achieve statistically valid results. Consequently, 156 scholars were recruited to participate in the study, ensuring robust representation and adequate statistical power. Data collection Fourteen proposed definitions of "scientist," curated from existing literature and expert contributions, were presented to the enrolled scholars for evaluation (Table 1). Each definition included a Likert scale with nine response options, ranging from "strongly agree" to "strongly disagree." Responses were systematically recorded and compiled in an Excel sheet for subsequent analysis. This process facilitated the systematic capture of scholarly consensus on each definition. Table 1. Respondent Agreement on Various 'Scientist' Definitions. Proposed Definitions Options A person studying or has expert knowledge of one or more natural or physical sciences. (Oxford Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree An expert who studies or works in one of the sciences. (Cambridge Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A person learned in science and especially natural science. (Merriam-Webster Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A scientist is someone who systematically gathers and uses research and evidence, to make hypotheses and test them, to gain and share understanding and knowledge. (Science Council) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A scientist is someone who has studied science and whose job is to teach or do research in science. (Collins Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree An expert in science, especially one of the physical or natural sciences. (Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A scientist is a person with some kind of knowledge or expertise in any of the sciences. (Vocabulary dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A person who is trained in a science and whose job involves doing scientific research or solving scientific problems. (Britannica Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A person who has studied science, especially one who is active in a particular field of investigation. (National Cancer Institute) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree Someone who works or is trained in science. (Longman Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A person whose profession is investigating in one of the natural sciences. (Your Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A person who is engaged in and has expert knowledge of a science. (Free Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree Someone whose job or education is about science. (LanGeek Dictionary) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree A scientist is a person who researches to advance knowledge in an area of the natural sciences. (Wikipedia) Strongly Disagree Moderately Disagree Disagree Slightly Disagree Undecided Slightly Agree Moderately Agree Agree Strongly Agree Data analysis The Content Validity Index (CVI) was employed to assess the relevance and agreement of the definitions. Definitions with a CVI below 0.70 were excluded, as they failed to meet the minimum threshold for consensus. Definitions with a CVI between 0.70 and 0.78 underwent a second round of evaluation, with refined wording sent back to the same scholars for further review. Definitions achieving a CVI above 0.78 were deemed sufficiently valid for inclusion in the final analysis [8]. These definitions formed the foundation for the development of a unified, consensus-based definition of "scientist." Ethical considerations Participation in the study was entirely voluntary, and all responses were anonymized to preserve participant confidentiality. Results Initially, out of the 14 proposed definitions of the term "scientist," six (42.9%) received a CVI score below the threshold of 0.70 and were consequently excluded from further consideration. In contrast, seven definitions (50.0%) demonstrated strong content validity with CVI scores equal to or exceeding 0.78 and were therefore retained for subsequent synthesis and analysis. Only one definition (7.1%) fell within the intermediate range, with a CVI between 0.70 and 0.78 (Table 2). Table 2. Comparison of Agreement and CVI Across Proposed Definitions of 'Scientist'. Proposed Definition Agree Disagree Undecided CVI Status Science Council Definition 146 8 2 0.94 Accepted Britannica Dictionary Definition 136 16 4 0.87 Accepted Cambridge Dictionary 130 20 6 0.83 Accepted Wikipedia Definition 125 27 4 0.80 Accepted Free Dictionary Definition 124 25 7 0.79 Accepted National Cancer Institute Definition 124 25 7 0.79 Accepted Collins Dictionary Definition 122 31 3 0.78 Accepted Oxford Dictionary Definition 120 31 5 0.77 Revised Longman Dictionary Definition 104 42 10 0.67 Excluded Your Dictionary Definition 103 48 5 0.66 Excluded Dictionary (generic) Definition 91 52 13 0.58 Excluded Vocabulary dictionary Definition 88 58 10 0.56 Excluded Merriam-Webster Dictionary Definition 81 64 11 0.52 Excluded LanGeek Dictionary Definition 79 67 10 0.51 Excluded CVI: Content Validity Index, CVI Thresholds: Accepted: ≥ 0.78, Revised: 0.70–0.78, Excluded: < 0.70 Through a rigorous, iterative evaluation process involving expert feedback, the definitions with the highest CVI scores (those above 0.78) were integrated and refined into two distinct, consensus-based definitions of the term "scientist." The first was a concise definition: “A scientist is a person who conducts research.” The second was a more comprehensive and elaborated definition: “A scientist is someone who systematically conducts or gathers and uses research to formulate hypotheses and test them, in order to gain and disseminate understanding and knowledge.” These two final definitions were subsequently circulated among the panel of scholars for a second round of evaluation, during which they were asked to rate the definitions for content validity. The short definition received a CVI of 0.82, while the more detailed definition attained a slightly higher CVI of 0.84, reflecting strong agreement among the experts. Although no additional formal qualitative feedback was solicited at this stage; minor wording adjustments were made based on informal suggestions received during this validation round. Discussion The role of a scientist extends far beyond the stereotypical image of an individual in a white coat working exclusively in a laboratory setting. Careers grounded in scientific expertise are remarkably diverse, encompassing domains such as research, education, industry, and regulatory affairs. The Science Council categorizes scientists into 10 different types, highlighting the diversity of scientific roles beyond the stereotypical lab-based researcher. It includes types such as experimental scientists, theoretical scientists, data scientists, and more, reflecting the broad spectrum of scientific work today [9]. Definitions of the term “scientist” vary, yet they generally converge on the principles of systematic inquiry, evidence-based investigation, and the pursuit of knowledge across various disciplines. For instance, the Oxford Advanced Learner’s Dictionary and the Britannica Dictionary emphasize formal training and research functions, typically within the natural sciences such as biology, chemistry, or physics [10,11]. In contrast, contemporary perspectives, such as those discussed by the American Association for the Advancement of Science in 2024, recognize a broader spectrum of scientific engagement, encompassing both professional researchers and individuals committed to understanding the world through observation, experimentation, and analysis [12]. In light of this diversity, the present study aimed to clarify and formalize the definition of a "scientist" through expert consensus. Two definitions were developed: a concise definition “A scientist is a person who conducts research”, and a comprehensive definition “A scientist is someone who systematically conducts or gathers and uses research to formulate hypotheses and test them, in order to gain and disseminate understanding and knowledge.” These definitions encapsulate the core activities and guiding principles of scientific inquiry, emphasizing both methodological rigor and the essential role of knowledge dissemination across disciplines. A key finding of this study lies in its recognition of the evolving tension between disciplinary specialization and the increasing importance of interdisciplinary collaboration. As highlighted in contemporary analyses of interdisciplinary research and development, scientists now frequently operate at the intersection of multiple fields, such as nanomedicine, where the diversity and dissimilarity of collaborators’ knowledge can significantly enhance research productivity [13]. The concise definition, "A scientist is a person who conducts research" captures this shift by avoiding constraints tied to specific disciplinary boundaries. In contrast, the more detailed definition explicitly incorporates the systematic formulation and testing of hypotheses, along with the dissemination of knowledge, thereby reinforcing the structured and communicative nature of scientific inquiry. These elements align closely with UNESCO’s 2019 call for stronger science-society engagement and underscore the ethical responsibilities inherent in modern scientific practice [14]. The study’s findings also contribute to ongoing debates surrounding professional identity within the scientific community. In contrast to regulated professions such as medicine, the absence of a universal credentialing system for scientists complicates formal recognition, particularly in non-academic and interdisciplinary contexts. This ambiguity is reflected in the National Cancer Institute’s pragmatic definition of a scientist, which emphasizes active participation in research rather than reliance on formal titles or qualifications [15]. By anchoring the term “scientist” in core research activities rather than occupational labels, the consensus-based definitions proposed in this study offer a more inclusive framework. This approach accommodates emerging roles in fields such as data science and applied research, thereby addressing the risk of under recognition in collaborative and cross-sector environments. The dual definitions, concise and comprehensive, offer flexibility for different contexts, a strategy aligned with the Science Council’s emphasis on methodological diversity [4]. The detailed definition’s focus on systematic inquiry and dissemination aligns with studies of interdisciplinary science, where “impassioned commitment” to shared goals drives innovation [13]. Simultaneously, the availability of a concise definition enhances clarity in public discourse and science communication, while the more detailed version provides the specificity necessary for institutional contexts such as policy development, research funding, and professional accreditation. Notably, the study’s findings also challenge enduring stereotypes of the “lone genius” scientist by highlighting the inherently collaborative and iterative nature of scientific practice. Contemporary frameworks, such as those emerging from computational biology, suggest that scientific identity is increasingly dynamic, pluralistic, and shaped by collective knowledge production [16]. The process undertaken in this study, involving successive refinement and expert validation of definitions, closely mirrors the recursive logic of the scientific method itself. This methodological alignment is particularly salient in fields like nutritional epidemiology, where the replication of findings remains a persistent challenge and iterative inquiry is essential for refining evidence [17]. Despite the methodological rigor and expert involvement, several limitations should be acknowledged. First, the study employed convenience sampling, which may introduce selection bias and limit the generalizability of the findings. Although participants were selected based on a minimum citation threshold to ensure scholarly expertise, this criterion may have inadvertently excluded emerging researchers or experts with significant practical contributions who have not yet achieved high citation metrics. Second, the use of an online survey format may have constrained participant engagement, as scholars with limited availability or preference for alternative formats may have been underrepresented. Additionally, response bias cannot be ruled out, as those with a particular interest in the topic or in defining scientific identity may have been more inclined to participate, potentially skewing the results. Future refinements of the definition should also consider voices from non-academic scientific contexts including those in industry, policy, and community-based science who are increasingly central to addressing complex global challenges. Conclusion By engaging experienced scholars across disciplines, this study establishes two validated definitions of “scientist” that emphasize systematic research activity and knowledge dissemination. These definitions offer a structured yet adaptable framework for understanding scientific identity, balancing clarity with flexibility. They help address the ambiguity surrounding the term “scientist,” providing a foundation for improved communication, interdisciplinary collaboration, and evidence-informed policy development. Importantly, they remain open to future refinement as scientific practice continues to evolve. Declarations Conflicts of interest: The authors have no conflicts of interest to disclose. Ethical approval: Not applicable. Patient consent (participation and publication): Not applicable. Funding: The present study received no financial support. Acknowledgements: None to be declared. Authors' contributions: JG, MM, SB, BS, VS, ASN, SHM, HAH, AGH, ADS, RAK, WRR, AB, GB, SS, SN, CJ, PL, MSS, ZK, MC, AM, SK, FCT, FB, FRK, MAM, AA, VK, DH, PM, VRM, MSA, EA, and RV were significant contributors to the conception of the study, voting for the items. FHK, BAA, and AMM were involved in the literature review, manuscript writing, and data analysis and interpretation. FHK and AMM Confirmation of the authenticity of all the raw data. All authors have read and approved the final version of the manuscript. Use of AI: ChatGPT-3.5 was used to assist in language editing and improving the clarity of the manuscript. All content was reviewed and verified by the authors. Authors are fully responsible for the entire content of their manuscript. Data availability statement: Not applicable.

Research of Protein-Protein Interaction (PPI) Network Alignment is playing an important role in understanding the crucial underlying biological knowledge such as functionally homologous proteins and conserved evolutionary pathways across different species. Existing methods of PPI network alignment often try to improve the coverage ratio of the alignment result by aligning all proteins from different species. However, there is a fundamental biological premise that needs to be considered carefully: not every protein in a species can, nor should, find its homologous proteins in other species. In this work, we propose a novel alignment method to map only those proteins with the most similarity throughout the PPI networks of multiple species. For the similarity features of the protein in the networks, we integrate both topological features with biological characteristics to provide enhanced supports for the alignment procedures. For topological features, we apply a representation learning method on the networks and generate a low dimensional vector embedding with its surrounding structural features for each protein. The topological similarity of proteins from different PPI networks can thus be transferred as the similarity of their corresponding vector representations, which provides a new way to comprehensively quantify the topological similarities between proteins. We also propose a new measure for the topological evaluation of the alignment results which better uncover the structural quality of the alignment across multiple networks. Both biological and topological evaluations on the alignment results of real datasets demonstrate our approach is promising and preferable against previous multiple alignment methods.

Network embedding, aiming to learn the low-dimensional representations of nodes in networks, is of paramount importance in many real applications. One basic requirement of network embedding is to preserve the structure and inherent properties of the networks. While previous network embedding methods primarily preserve the microscopic structure, such as the first- and second-order proximities of nodes, the mesoscopic community structure, which is one of the most prominent feature of networks, is largely ignored. In this paper, we propose a novel Modularized Nonnegative Matrix Factorization (M-NMF) model to incorporate the community structure into network embedding. We exploit the consensus relationship between the representations of nodes and community structure, and then jointly optimize NMF based representation learning model and modularity based community detection model in a unified framework, which enables the learned representations of nodes to preserve both of the microscopic and community structures. We also provide efficient updating rules to infer the parameters of our model, together with the correctness and convergence guarantees. Extensive experimental results on a variety of real-world networks show the superior performance of the proposed method over the state-of-the-arts.

There was proximity between natural and social science in the history.It couldn't be say that, this proximity continues now.Because sciencis were divided into specialist areas.Scholar who lived even in the middle age or new ages could have profound knowledge in the field of social sciences and natural sciences.While explaining social science scholars that could benefit from the basic information in the field of natürel science.Social science concepts can be intangible.At this point it may be necessary to explain some concepts concretization.It makes it easier to understand and permanence of the affects if we tell you a thing or terminology by concretizing.For this, social science was used from natural science terminology that time to time.We could give some example for this hypothesis.An example of relationship between sociology and biology or political science and sociology-biology-physics which was particularly marked.Especially middle age and eastern scientists have put forward opinions about relationship between social and natural science.For example, Ibn Khaldun, likened the individual to the state.These smilarities has established relevance between science of political and science of biology.On the other side the west scholars have demonstrated many examples of ideas or hypotheses that relationship in new age.

Network alignment is the matching of two networks with corresponding nodes that belong to the same user or entity. The most common application is to analyze which accounts belong to the same user in two social networks. Most of existing techniques rely on matrix factorization so that they cannot be scaled to large-scale networks, are constrained by strict constraints, and cannot learn node embedding without a training set. In this paper, we propose an unsupervised network alignment model based on multi-level graph attention networks. The model uses multi-level graph attention network to learn the embedded representation of nodes, satisfying attribute and structure constraints of alignment. Augmented learning process is proposed to simulate attribute noise and structural noise to improve adaptability of the model. Extensive experiments on real datasets show that the proposed model performs better than the state-of-the-art network alignment model. We also demonstrate the robustness of the proposed model.

The view that social sciences are very different from other forms of science is common, hence the need to develop special methods. In section 1.1 it is argued that, although there are some important differences, modern developments both in social sciences and in what are called here natural sciences tend to reduce them, pointing towards a common methodological framework. This idea serves as an introduction to further sections where the scientific method for the social sciences is outlined, and then is shown the role that models play in it. Natural and social sciences: the hypothesis of convergency Traditionally, social sciences have been considered a very special form of science, different from biology, physics or natural sciences in general. Many arguments are put forward to support this distinction. By contrast, social phenomena are considered to be ever:changing. It may well be that in natural sciences theories themselves change and old ones are replaced by new ones, but once adopted, they are considered as permanent. If social phenomena are ever:changing, social theories may be valid only for a short period of time; that is, they have a particular historical reference. Because it is generally assumed that natural phenomena are permanent, theories that successfully explain their past behaviour have strong predictive power. In social sciences, even if a theory has been very successful in explaining a phenomenon that occurred in the past, it can only provide predictions as long as the historical conditions prevail. Another difference that is commonly pointed out is that natural phenomena can be reproduced in the controlled conditions of a laboratory, which isolates them from unwanted externalities. The use of experiments has become a key element in the creation of knowledge in natural sciences.