The purpose of the paper is to present a model of factors affecting the successful project implementation by introducing agility and artificial intelligence to increase the company’s competitiveness. In the model, the multidimensional constructs describing the implementation of an agile work environment and artificial intelligence technologies and tools were developed. These multidimensional constructs are agile work environment, agile leadership, agile team skills and capabilities, improving the work of the leader in the project, adopting AI technologies in the project, and using AI solutions in a project. Their impact on successful project implementation and on the company competitiveness was tested. The fundamental reason for conducting this research and developing the model is to enhance the understanding of factors that contribute to the successful implementation of projects and to increase a company’s competitiveness. Our developed model encompasses multidimensional constructs that describe the agile work environment and the utilization of AI technologies. By examining the impact of these constructs on both successful project implementation and company competitiveness, we aimed to establish a comprehensive framework that captures the relationship between agility, AI, and successful project implementation. This model serves as a valuable tool for companies seeking to improve their project implementation processes and gain a competitive edge in the market. The research was based on a sample of 473 managers/owners in medium-sized and large companies. Structural equation modeling was used to test the hypotheses. In today’s turbulent environment, the results will help develop guidelines for a successful combination of agile business practices and artificial intelligence to achieve successful project implementation, increasing a company’s competitiveness.
{"title":"Artificial Intelligence and Agility-Based Model for Successful Project Implementation and Company Competitiveness","authors":"P. Tominc, D. Oreški, M. Rožman","doi":"10.3390/info14060337","DOIUrl":"https://doi.org/10.3390/info14060337","url":null,"abstract":"The purpose of the paper is to present a model of factors affecting the successful project implementation by introducing agility and artificial intelligence to increase the company’s competitiveness. In the model, the multidimensional constructs describing the implementation of an agile work environment and artificial intelligence technologies and tools were developed. These multidimensional constructs are agile work environment, agile leadership, agile team skills and capabilities, improving the work of the leader in the project, adopting AI technologies in the project, and using AI solutions in a project. Their impact on successful project implementation and on the company competitiveness was tested. The fundamental reason for conducting this research and developing the model is to enhance the understanding of factors that contribute to the successful implementation of projects and to increase a company’s competitiveness. Our developed model encompasses multidimensional constructs that describe the agile work environment and the utilization of AI technologies. By examining the impact of these constructs on both successful project implementation and company competitiveness, we aimed to establish a comprehensive framework that captures the relationship between agility, AI, and successful project implementation. This model serves as a valuable tool for companies seeking to improve their project implementation processes and gain a competitive edge in the market. The research was based on a sample of 473 managers/owners in medium-sized and large companies. Structural equation modeling was used to test the hypotheses. In today’s turbulent environment, the results will help develop guidelines for a successful combination of agile business practices and artificial intelligence to achieve successful project implementation, increasing a company’s competitiveness.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79693439","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
(1) Background: The prevalence of uncontrolled hypertension is rising all across the world, making it a concern for public health. The usage of mobile health applications has resulted in a number of positive outcomes for the management and control of hypertension. (2) Objective: The study’s primary goal is to explain the steps to create a hypertension application (app) that considers cultural and social standards in Saudi Arabia, motivational features, and the needs of male and female Saudi citizens. (3) Methods: This study reports the emerged features and content needed to be adapted or developed in health apps for hypertension patients during an interactive qualitative analysis focus group activity with (n = 5) experts from the Saudi Ministry of Health. A gap analysis was conducted to develop an app based on a deep understanding of user needs with a patient-centred approach. (4) Results: Based on the participant’s reviews in this study, the app was easy to use and can help Saudi patients to control their hypertension, the design was interactive, motivational features are user-friendly, and there is a need to consider other platforms such as Android and Blackberry in a future version. (5) Conclusions: Mobile health apps can help Saudis change their unhealthy lifestyles. Target users, usability, motivational features, and social and cultural standards must be considered to meet the app’s aim.
{"title":"Enabled Artificial Intelligence (AI) to Develop Sehhaty Wa Daghty App of Self-Management for Saudi Patients with Hypertension: A Qualitative Study","authors":"Adel Alzahrani, V. Gay, Ryan Alturki","doi":"10.3390/info14060334","DOIUrl":"https://doi.org/10.3390/info14060334","url":null,"abstract":"(1) Background: The prevalence of uncontrolled hypertension is rising all across the world, making it a concern for public health. The usage of mobile health applications has resulted in a number of positive outcomes for the management and control of hypertension. (2) Objective: The study’s primary goal is to explain the steps to create a hypertension application (app) that considers cultural and social standards in Saudi Arabia, motivational features, and the needs of male and female Saudi citizens. (3) Methods: This study reports the emerged features and content needed to be adapted or developed in health apps for hypertension patients during an interactive qualitative analysis focus group activity with (n = 5) experts from the Saudi Ministry of Health. A gap analysis was conducted to develop an app based on a deep understanding of user needs with a patient-centred approach. (4) Results: Based on the participant’s reviews in this study, the app was easy to use and can help Saudi patients to control their hypertension, the design was interactive, motivational features are user-friendly, and there is a need to consider other platforms such as Android and Blackberry in a future version. (5) Conclusions: Mobile health apps can help Saudis change their unhealthy lifestyles. Target users, usability, motivational features, and social and cultural standards must be considered to meet the app’s aim.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88290404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Canonical ensembles of walks in a finite connected graph assign the properly normalized probability distributions to all nodes, subgraphs, and nodal subsets of the graph at all time and connectivity scales of the diffusion process. The probabilistic description of graphs allows for introducing the quantitative measures of navigability through the graph, walkability of individual paths, and mutual perspicacity of the different modes of the (diffusion) processes. The application of information theory methods to problems about graphs, in contrast to geometric, combinatoric, algorithmic, and algebraic approaches, can be called information graph theory. As it involves evaluating communication efficiency between individual systems’ units at different time and connectivity scales, information graph theory is in demand for a wide range of applications, such as designing network-on-chip architecture and engineering urban morphology within the concept of the smart city.
{"title":"Navigability, Walkability, and Perspicacity Associated with Canonical Ensembles of Walks in Finite Connected Undirected Graphs - Toward Information Graph Theory","authors":"D. Volchenkov","doi":"10.3390/info14060338","DOIUrl":"https://doi.org/10.3390/info14060338","url":null,"abstract":"Canonical ensembles of walks in a finite connected graph assign the properly normalized probability distributions to all nodes, subgraphs, and nodal subsets of the graph at all time and connectivity scales of the diffusion process. The probabilistic description of graphs allows for introducing the quantitative measures of navigability through the graph, walkability of individual paths, and mutual perspicacity of the different modes of the (diffusion) processes. The application of information theory methods to problems about graphs, in contrast to geometric, combinatoric, algorithmic, and algebraic approaches, can be called information graph theory. As it involves evaluating communication efficiency between individual systems’ units at different time and connectivity scales, information graph theory is in demand for a wide range of applications, such as designing network-on-chip architecture and engineering urban morphology within the concept of the smart city.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75455905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Knowledge graphs are graph-based data models which can represent real-time data that is constantly growing with the addition of new information. The question-answering systems over knowledge graphs (KGQA) retrieve answers to a natural language question from the knowledge graph. Most existing KGQA systems use static knowledge bases for offline training. After deployment, they fail to learn from unseen new entities added to the graph. There is a need for dynamic algorithms which can adapt to the evolving graphs and give interpretable results. In this research work, we propose using new auction algorithms for question answering over knowledge graphs. These algorithms can adapt to changing environments in real-time, making them suitable for offline and online training. An auction algorithm computes paths connecting an origin node to one or more destination nodes in a directed graph and uses node prices to guide the search for the path. The prices are initially assigned arbitrarily and updated dynamically based on defined rules. The algorithm navigates the graph from the high-price to the low-price nodes. When new nodes and edges are dynamically added or removed in an evolving knowledge graph, the algorithm can adapt by reusing the prices of existing nodes and assigning arbitrary prices to the new nodes. For subsequent related searches, the “learned” prices provide the means to “transfer knowledge” and act as a “guide”: to steer it toward the lower-priced nodes. Our approach reduces the search computational effort by 60% in our experiments, thus making the algorithm computationally efficient. The resulting path given by the algorithm can be mapped to the attributes of entities and relations in knowledge graphs to provide an explainable answer to the query. We discuss some applications for which our method can be used.
{"title":"Auction-Based Learning for Question Answering over Knowledge Graphs","authors":"Garima Agrawal, D. Bertsekas, Huan Liu","doi":"10.3390/info14060336","DOIUrl":"https://doi.org/10.3390/info14060336","url":null,"abstract":"Knowledge graphs are graph-based data models which can represent real-time data that is constantly growing with the addition of new information. The question-answering systems over knowledge graphs (KGQA) retrieve answers to a natural language question from the knowledge graph. Most existing KGQA systems use static knowledge bases for offline training. After deployment, they fail to learn from unseen new entities added to the graph. There is a need for dynamic algorithms which can adapt to the evolving graphs and give interpretable results. In this research work, we propose using new auction algorithms for question answering over knowledge graphs. These algorithms can adapt to changing environments in real-time, making them suitable for offline and online training. An auction algorithm computes paths connecting an origin node to one or more destination nodes in a directed graph and uses node prices to guide the search for the path. The prices are initially assigned arbitrarily and updated dynamically based on defined rules. The algorithm navigates the graph from the high-price to the low-price nodes. When new nodes and edges are dynamically added or removed in an evolving knowledge graph, the algorithm can adapt by reusing the prices of existing nodes and assigning arbitrary prices to the new nodes. For subsequent related searches, the “learned” prices provide the means to “transfer knowledge” and act as a “guide”: to steer it toward the lower-priced nodes. Our approach reduces the search computational effort by 60% in our experiments, thus making the algorithm computationally efficient. The resulting path given by the algorithm can be mapped to the attributes of entities and relations in knowledge graphs to provide an explainable answer to the query. We discuss some applications for which our method can be used.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82826402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Federico Manuri, Federico Decataldo, A. Sanna, Paolo Brizzi
Virtual environments have been widely adopted for design and training tasks in the industrial domain. Low-cost automation (LCA) is a technology that automatizes some activities using mostly standard automation mechanisms available off the shelf. However, LCA systems should adapt to existing standard production lines and workstations. Thus, workers must customize standard LCA templates and perform adaptation and customization steps. This activity can be very time consuming with physical LCA systems, and in case of errors, it may be necessary to rebuild many parts from scratch. Thus, LCA systems would greatly benefit from a design and prototyping step experienced in a virtual simulation environment. An immersive virtual reality (IVR) application for rapid and easy prototyping of LCA solutions has been investigated in previous work; the assessment of the system usability proved that the users highly appreciated the proposed solutions. This research explores further improvements to exploit the existing IVR application as a training tool for LCA prototyping trainees. The proposed application now provides users with two different interaction paradigms based on the VIVE controllers and the Manus Prime II data gloves. The application’s interface has been revised to allow a proper comparison of the two interaction models. The two interfaces have been compared, involving 12 participants in an LCA building task. The System Usability Scale (SUS) and the NASA Task Load Index (TLX) questionnaires have been used to assess the usability and workload of the two solutions.
虚拟环境已被广泛应用于工业领域的设计和培训任务。低成本自动化(LCA)是一种使用现成的标准自动化机制来自动化某些活动的技术。然而，LCA系统应该适应现有的标准生产线和工作站。因此，工作人员必须自定义标准LCA模板，并执行调整和自定义步骤。对于物理LCA系统，此活动可能非常耗时，并且在出现错误的情况下，可能需要从头开始重新构建许多部件。因此，LCA系统将从虚拟仿真环境中的设计和原型步骤中受益匪浅。一种沉浸式虚拟现实(IVR)应用程序，用于快速和简单的LCA解决方案原型设计，已经在以前的工作中进行了研究;系统的可用性评估表明，用户对所提出的解决方案高度赞赏。本研究探索了进一步的改进，以利用现有的IVR应用程序作为LCA原型学员的培训工具。该应用程序现在为用户提供了两种不同的基于VIVE控制器和Manus Prime II数据手套的交互模式。该应用程序的接口已被修改，以允许对两种交互模型进行适当的比较。在LCA构建任务中，对这两个接口进行了比较，涉及12个参与者。使用系统可用性量表(SUS)和NASA任务负载指数(TLX)问卷来评估这两种解决方案的可用性和工作量。
{"title":"A Comparison of Two Interaction Paradigms for Training Low Cost Automation Assembly in Virtual Environments","authors":"Federico Manuri, Federico Decataldo, A. Sanna, Paolo Brizzi","doi":"10.3390/info14060340","DOIUrl":"https://doi.org/10.3390/info14060340","url":null,"abstract":"Virtual environments have been widely adopted for design and training tasks in the industrial domain. Low-cost automation (LCA) is a technology that automatizes some activities using mostly standard automation mechanisms available off the shelf. However, LCA systems should adapt to existing standard production lines and workstations. Thus, workers must customize standard LCA templates and perform adaptation and customization steps. This activity can be very time consuming with physical LCA systems, and in case of errors, it may be necessary to rebuild many parts from scratch. Thus, LCA systems would greatly benefit from a design and prototyping step experienced in a virtual simulation environment. An immersive virtual reality (IVR) application for rapid and easy prototyping of LCA solutions has been investigated in previous work; the assessment of the system usability proved that the users highly appreciated the proposed solutions. This research explores further improvements to exploit the existing IVR application as a training tool for LCA prototyping trainees. The proposed application now provides users with two different interaction paradigms based on the VIVE controllers and the Manus Prime II data gloves. The application’s interface has been revised to allow a proper comparison of the two interaction models. The two interfaces have been compared, involving 12 participants in an LCA building task. The System Usability Scale (SUS) and the NASA Task Load Index (TLX) questionnaires have been used to assess the usability and workload of the two solutions.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91379385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rajagopal Peesapati, Y. K. Nayak, Swati K. Warungase, S. Salkuti
The rapid growth in greenhouse gases (GHGs), the lack of electricity production, and an ever-increasing demand for electrical energy requires an optimal reduction in coal-fired thermal generating units (CFTGU) with the aim of minimizing fuel costs and emissions. Previous approaches have been unable to deal with such problems due to the non-convexity of realistic scenarios and confined optimum convergence. Instead, meta-heuristic techniques have gained more attention in order to deal with such constrained static/dynamic economic emission load dispatch (ELD/DEELD) problems, due to their flexibility and derivative-free structures. Hence, in this work, the elephant herd optimization (EHO) technique is proposed in order to solve constrained non-convex static and dynamic ELD problems in the power system. The proposed EHO algorithm is a nature-inspired technique that utilizes a new separation method and elitism strategy in order to retain the diversity of the population and to ensure that the fittest individuals are retained in the next generation. The current approach can be implemented to minimize both the fuel and emission cost functions of the CFTGUs subject to power balance constraints, active power generation limits, and ramp rate limits in the system. Three test systems involving 6, 10, and 40 units were utilized to demonstrate the effectiveness and practical feasibility of the proposed algorithm. Numerical results indicate that the proposed EHO algorithm exhibits better performance in most of the test cases as compared to recent existing algorithms when applied to the static and dynamic ELD issue, demonstrating its superiority and practicability.
{"title":"Constrained Static/Dynamic Economic Emission Load Dispatch Using Elephant Herd Optimization","authors":"Rajagopal Peesapati, Y. K. Nayak, Swati K. Warungase, S. Salkuti","doi":"10.3390/info14060339","DOIUrl":"https://doi.org/10.3390/info14060339","url":null,"abstract":"The rapid growth in greenhouse gases (GHGs), the lack of electricity production, and an ever-increasing demand for electrical energy requires an optimal reduction in coal-fired thermal generating units (CFTGU) with the aim of minimizing fuel costs and emissions. Previous approaches have been unable to deal with such problems due to the non-convexity of realistic scenarios and confined optimum convergence. Instead, meta-heuristic techniques have gained more attention in order to deal with such constrained static/dynamic economic emission load dispatch (ELD/DEELD) problems, due to their flexibility and derivative-free structures. Hence, in this work, the elephant herd optimization (EHO) technique is proposed in order to solve constrained non-convex static and dynamic ELD problems in the power system. The proposed EHO algorithm is a nature-inspired technique that utilizes a new separation method and elitism strategy in order to retain the diversity of the population and to ensure that the fittest individuals are retained in the next generation. The current approach can be implemented to minimize both the fuel and emission cost functions of the CFTGUs subject to power balance constraints, active power generation limits, and ramp rate limits in the system. Three test systems involving 6, 10, and 40 units were utilized to demonstrate the effectiveness and practical feasibility of the proposed algorithm. Numerical results indicate that the proposed EHO algorithm exhibits better performance in most of the test cases as compared to recent existing algorithms when applied to the static and dynamic ELD issue, demonstrating its superiority and practicability.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86437848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Enrico De Santis, A. Martino, Francesca Ronci, A. Rizzi
On 24 February 2022, the invasion of Ukraine by Russian troops began, starting a dramatic conflict. As in all modern conflicts, the battlefield is both real and virtual. Social networks have had peaks in use and many scholars have seen a strong risk of disinformation. In this study, through an unsupervised topic tracking system implemented with Natural Language Processing and graph-based techniques framed within a biological metaphor, the Italian social context is analyzed, in particular, by processing data from Twitter (texts and metadata) captured during the first month of the war. The system, improved if compared to previous versions, has proved to be effective in highlighting the emerging topics, all the main events and any links between them.
{"title":"An Unsupervised Graph-Based Approach for Detecting Relevant Topics: A Case Study on the Italian Twitter Cohort during the Russia-Ukraine Conflict","authors":"Enrico De Santis, A. Martino, Francesca Ronci, A. Rizzi","doi":"10.3390/info14060330","DOIUrl":"https://doi.org/10.3390/info14060330","url":null,"abstract":"On 24 February 2022, the invasion of Ukraine by Russian troops began, starting a dramatic conflict. As in all modern conflicts, the battlefield is both real and virtual. Social networks have had peaks in use and many scholars have seen a strong risk of disinformation. In this study, through an unsupervised topic tracking system implemented with Natural Language Processing and graph-based techniques framed within a biological metaphor, the Italian social context is analyzed, in particular, by processing data from Twitter (texts and metadata) captured during the first month of the war. The system, improved if compared to previous versions, has proved to be effective in highlighting the emerging topics, all the main events and any links between them.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80494177","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Sobhi, M. Reshadi, Nick Zarft, Albert Terheide, Scott Dick
Electric induction motors are one of the most important and widely used classes of machines in modern industry. Large motors, which are commonly process-critical, will usually have built-in condition-monitoring systems to facilitate preventive maintenance and fault detection. Such capabilities are usually not cost-effective for small (under ten horsepower) motors, as they are inexpensive to replace. However, large industrial sites may use hundreds of these small motors, often to drive cooling fans or lubrication pumps for larger machines. Multiple small motors may further be assigned to a single electrical circuit, meaning a failure in one could damage other motors on that circuit. There is thus a need for condition monitoring of aggregations of small motors. We report on an ongoing project to develop a machine-learning-based solution for fault detection in multiple small electric motors. Shallow and deep learning approaches to this problem are investigated and compared, with a hybrid deep/shallow system ultimately being the most effective.
{"title":"Condition Monitoring and Fault Detection in Small Induction Motors Using Machine Learning Algorithms","authors":"S. Sobhi, M. Reshadi, Nick Zarft, Albert Terheide, Scott Dick","doi":"10.3390/info14060329","DOIUrl":"https://doi.org/10.3390/info14060329","url":null,"abstract":"Electric induction motors are one of the most important and widely used classes of machines in modern industry. Large motors, which are commonly process-critical, will usually have built-in condition-monitoring systems to facilitate preventive maintenance and fault detection. Such capabilities are usually not cost-effective for small (under ten horsepower) motors, as they are inexpensive to replace. However, large industrial sites may use hundreds of these small motors, often to drive cooling fans or lubrication pumps for larger machines. Multiple small motors may further be assigned to a single electrical circuit, meaning a failure in one could damage other motors on that circuit. There is thus a need for condition monitoring of aggregations of small motors. We report on an ongoing project to develop a machine-learning-based solution for fault detection in multiple small electric motors. Shallow and deep learning approaches to this problem are investigated and compared, with a hybrid deep/shallow system ultimately being the most effective.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79434490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Community detection in dynamic networks is a challenging research problem. One of the main obstacles is the stability issues that arise during the evolution of communities. In dynamic networks, new communities may emerge and existing communities may disappear, grow, or shrink. As a result, a community can evolve into a completely different one, making it difficult to track its evolution (this is known as the drifting/identity problem). In this paper, we focused on the evolution of a single community. Our aim was to identify the community that contains a particularly important node, called the anchor, and to track its evolution over time. In this way, we circumvented the identity problem by allowing the anchor to define the core of the relevant community. We proposed a framework that tracks the evolution of the community defined by the anchor and verified its efficiency and effectiveness through experimental evaluation.
{"title":"Local Community Detection in Graph Streams with Anchors","authors":"Konstantinos Christopoulos, Georgia Baltsou, Konstantinos Tsichlas","doi":"10.3390/info14060332","DOIUrl":"https://doi.org/10.3390/info14060332","url":null,"abstract":"Community detection in dynamic networks is a challenging research problem. One of the main obstacles is the stability issues that arise during the evolution of communities. In dynamic networks, new communities may emerge and existing communities may disappear, grow, or shrink. As a result, a community can evolve into a completely different one, making it difficult to track its evolution (this is known as the drifting/identity problem). In this paper, we focused on the evolution of a single community. Our aim was to identify the community that contains a particularly important node, called the anchor, and to track its evolution over time. In this way, we circumvented the identity problem by allowing the anchor to define the core of the relevant community. We proposed a framework that tracks the evolution of the community defined by the anchor and verified its efficiency and effectiveness through experimental evaluation.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81219163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Video question answering (QA) is a cross-modal task that requires understanding the video content to answer questions. Current techniques address this challenge by employing stacked modules, such as attention mechanisms and graph convolutional networks. These methods reason about the semantics of video features and their interaction with text-based questions, yielding excellent results. However, these approaches often learn and fuse features representing different aspects of the video separately, neglecting the intra-interaction and overlooking the latent complex correlations between the extracted features. Additionally, the stacking of modules introduces a large number of parameters, making model training more challenging. To address these issues, we propose a novel multimodal knowledge distillation method that leverages the strengths of knowledge distillation for model compression and feature enhancement. Specifically, the fused features in the larger teacher model are distilled into knowledge, which guides the learning of appearance and motion features in the smaller student model. By incorporating cross-modal information in the early stages, the appearance and motion features can discover their related and complementary potential relationships, thus improving the overall model performance. Despite its simplicity, our extensive experiments on the widely used video QA datasets, MSVD-QA and MSRVTT-QA, demonstrate clear performance improvements over prior methods. These results validate the effectiveness of the proposed knowledge distillation approach.
{"title":"A Video Question Answering Model Based on Knowledge Distillation","authors":"Zhuang Shao, Jiahui Wan, Linlin Zong","doi":"10.3390/info14060328","DOIUrl":"https://doi.org/10.3390/info14060328","url":null,"abstract":"Video question answering (QA) is a cross-modal task that requires understanding the video content to answer questions. Current techniques address this challenge by employing stacked modules, such as attention mechanisms and graph convolutional networks. These methods reason about the semantics of video features and their interaction with text-based questions, yielding excellent results. However, these approaches often learn and fuse features representing different aspects of the video separately, neglecting the intra-interaction and overlooking the latent complex correlations between the extracted features. Additionally, the stacking of modules introduces a large number of parameters, making model training more challenging. To address these issues, we propose a novel multimodal knowledge distillation method that leverages the strengths of knowledge distillation for model compression and feature enhancement. Specifically, the fused features in the larger teacher model are distilled into knowledge, which guides the learning of appearance and motion features in the smaller student model. By incorporating cross-modal information in the early stages, the appearance and motion features can discover their related and complementary potential relationships, thus improving the overall model performance. Despite its simplicity, our extensive experiments on the widely used video QA datasets, MSVD-QA and MSRVTT-QA, demonstrate clear performance improvements over prior methods. These results validate the effectiveness of the proposed knowledge distillation approach.","PeriodicalId":13622,"journal":{"name":"Inf. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75870567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：