Data & Knowledge Engineering最新文献

{"title":"Recognition algorithm for cross-texting in text chat conversations","authors":"Da-Young Lee,&nbsp;Hwan-Gue Cho","doi":"10.1016/j.datak.2023.102261","DOIUrl":"10.1016/j.datak.2023.102261","url":null,"abstract":"<div><p>As the development of the Internet and IT technology, short-text based communication is so popular compared with voice based one. Chat-based communication enables rapid, short and massive exchange of message with many people, creates new social problems. ‘Cross-texting’ is one of them. It refers to accidentally sending a text to an unintended person during the concurrent conversations with separated multiple people. Cross-texting would be a serious problem in languages where respectful expressions are required. As text-based communication is getting popular, it is a crucial work to prevent cross-texting by detecting it in advance in languages with honorifics expression such as Korean. In this paper, we proposed two methods detecting a cross-text using a deep learning model<span>. The first model is the formal feature vector, which models dialog by explicitly defining the politeness and completeness features. The second one is the grpah2vec based ChatGram-net model, which models the dialog based on the syllable occurrence relationship. To evaluate the detection performance, we suggest a generating method for cross-text datasets from a actual messenger corpus. In experiment we show that both proposed models detected cross-text effectively, and exceeded the performance of the baseline models.</span></p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"150 ","pages":"Article 102261"},"PeriodicalIF":2.5,"publicationDate":"2023-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138576764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards deep understanding of graph convolutional networks for relation extraction","authors":"Tao Wu ,&nbsp;Xiaolin You ,&nbsp;Xingping Xian ,&nbsp;Xiao Pu ,&nbsp;Shaojie Qiao ,&nbsp;Chao Wang","doi":"10.1016/j.datak.2023.102265","DOIUrl":"10.1016/j.datak.2023.102265","url":null,"abstract":"<div><p><span><span>Relation extraction aims at identifying semantic relations between pairs of named entities from unstructured texts and is considered an essential prerequisite for many downstream tasks in </span>natural language processing (NLP). Owing to the ability in expressing complex relationships and </span>interdependency<span><span><span>, graph neural networks<span> (GNNs) have been gradually used to solve the relation extraction problem and have achieved state-of-the-art results. However, the designs of GNN-based relation extraction methods are mostly based on empirical intuition, heuristic, and experimental trial-and-error. A clear understanding of why and how GNNs perform well in relation extraction tasks is lacking. In this study, we investigate three well-known GNN-based relation extraction models, CGCN, AGGCN, and SGCN, and aim to understand the underlying mechanisms of the extractions. In particular, we provide a </span></span>visual analytic to reveal the dynamics of the models and provide insight into the function of intermediate </span>convolutional layers. We determine that entities, particularly subjects and objects in them, are more important features than other words for relation extraction tasks. With various masking strategies, the significance of entity type to relation extraction is recognized. Then, from the perspective of the model architecture, we find that graph structure modeling and aggregation mechanisms in GCN do not significantly affect the performance improvement of GCN-based relation extraction models. The above findings are of great significance in promoting the development of GNNs. Based on these findings, an engineering oriented MLP-based GNN relation extraction model is proposed to achieve a comparable performance and greater efficiency.</span></p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"149 ","pages":"Article 102265"},"PeriodicalIF":2.5,"publicationDate":"2023-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138546596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Generating psychological analysis tables for children's drawings using deep learning","authors":"Moonyoung Lee ,&nbsp;Youngho Kim ,&nbsp;Young-Kuk Kim","doi":"10.1016/j.datak.2023.102266","DOIUrl":"10.1016/j.datak.2023.102266","url":null,"abstract":"<div><p>The usefulness of drawing-based psychological testing has been demonstrated in a variety of studies. By using the familiar medium of drawing, drawing-based psychological testing can be applied to a wide range of age groups and is particularly effective with children who have difficulty expressing themselves verbally. Drawing tests are usually implemented face-to-face, requiring specialized counseling staff, and can be time-consuming and expensive to apply to large numbers of children. These problems seem to be solved by applying highly developed artificial intelligence<span> techniques. If artificial intelligence (AI) can analyze children's drawings and perform psychological analysis, it will be possible to use it as a service and take tests online or through smartphones. There have been various attempts to automate the drawing of psychological tests by utilizing deep learning technology to process images. Previous studies using classification have been limited in their ability to extract structural information. In this paper, we analyze the House-Tree-Person Test (HTP), one of the drawing psychological tests widely used in clinical practice, by utilizing object detection technology that can extract more diverse information from images. In addition, we extend the existing research that has been limited to the extraction of relatively simple psychological features and generate a psychological analysis table based on the extracted features that can be used to assist experts in the process of psychological testing. Our research findings indicate that the object detection performance achieves a mean Average Precision (mAP) of approximately 92.6∼94.1 %, and the average accuracy of the psychological analysis table is 94.4 %.</span></p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"149 ","pages":"Article 102266"},"PeriodicalIF":2.5,"publicationDate":"2023-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138546528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Blockchain-based ontology driven reference framework for security risk management","authors":"Mubashar Iqbal ,&nbsp;Aleksandr Kormiltsyn ,&nbsp;Vimal Dwivedi ,&nbsp;Raimundas Matulevičius","doi":"10.1016/j.datak.2023.102257","DOIUrl":"10.1016/j.datak.2023.102257","url":null,"abstract":"<div><p>Security risk management<span><span> (SRM) is crucial for protecting valuable assets from malicious harm. While blockchain technology has been proposed to mitigate security threats in traditional applications, it is not a perfect solution, and its security threats must be managed. This paper addresses the research problem of having no unified and formal knowledge models to support the SRM of traditional applications using blockchain and the SRM of blockchain-based applications. In accordance with this, we present a blockchain-based reference model (BbRM) and an ontology driven reference framework (OntReF) for the SRM of traditional and blockchain-based applications. The BbRM consolidates security threats of traditional and blockchain-based applications, structured following the SRM domain model and offers guidance for creating the OntReF using the domain model. OntReF is grounded on unified foundational ontology (UFO) and provides semantic interoperability and supporting the dynamic knowledge representation and </span>instantiation of information security knowledge for the SRM. Our evaluation approaches demonstrate that OntReF is practical to use.</span></p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"149 ","pages":"Article 102257"},"PeriodicalIF":2.5,"publicationDate":"2023-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138534223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated detection and localization of concept drifts in process mining with batch and stream trace clustering support","authors":"Rafael Gaspar de Sousa ,&nbsp;Antonio Carlos Meira Neto ,&nbsp;Marcelo Fantinato ,&nbsp;Sarajane Marques Peres ,&nbsp;Hajo Alexander Reijers","doi":"10.1016/j.datak.2023.102253","DOIUrl":"10.1016/j.datak.2023.102253","url":null,"abstract":"<div><p><span>Process mining can help organizations by extracting knowledge from event logs. However, process mining techniques often assume business processes are stationary, while actual business processes are constantly subject to change because of the complexity of organizations and their external environment. Thus, addressing process changes over time – known as </span><em>concept drifts</em><span><span><span><span> – allows for a better understanding of process behavior and can provide a competitive edge for organizations, especially in an online data stream scenario. Current approaches to handling process concept drift focus primarily on detecting and locating concept drifts, often through an integrated, albeit offline, approach. However, part of these integrated approaches rely on complex </span>data structures<span> related to tree-based process models, usually discovered through algorithms whose results are influenced by specific heuristic rules. Moreover, most of the proposed approaches have not been tested on public true concept drift-labeled event logs commonly used as benchmark, making comparative analysis difficult. In this article, we propose an online approach to detect and localize concept drifts in an integrated way using batch and stream trace clustering support. In our approach, cluster models provide input information for both concept drift detection and </span></span>localization methods. Each cluster abstracts a behavior profile underlying the process and reveals </span>descriptive information about the discovered concept drifts. Experiments with benchmark synthetic event logs with different control-flow changes, as well as with real-world event logs, showed that our approach, when relying on the same clustering model, is competitive in relation to baselines concept drift detection method. In addition, the experiment showed our approach is able to correctly locate the concept drifts detected and allows the analysis of such concept drifts through different process behavior profiles.</span></p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"149 ","pages":"Article 102253"},"PeriodicalIF":2.5,"publicationDate":"2023-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138534225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial for VSI:NLDB-saarbruecken-2021","authors":"Helmut Horacek ,&nbsp;Epaminondas Kapetanios ,&nbsp;Elisabeth Metais ,&nbsp;Farid Meziane","doi":"10.1016/j.datak.2023.102259","DOIUrl":"https://doi.org/10.1016/j.datak.2023.102259","url":null,"abstract":"","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"149 ","pages":"Article 102259"},"PeriodicalIF":2.5,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138474745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DeepScraper: A complete and efficient tweet scraping method using authenticated multiprocessing","authors":"Jaebeom You ,&nbsp;Kisung Lee ,&nbsp;Hyuk-Yoon Kwon","doi":"10.1016/j.datak.2023.102260","DOIUrl":"10.1016/j.datak.2023.102260","url":null,"abstract":"<div><p>In this paper, we propose a scraping method for collecting tweets, which we call <em>DeepScraper</em><span>. DeepScraper provides the complete scraping for the entire tweets written by a certain group of users or them containing search keywords<span> with a fast speed. To improve the crawling speed of DeepScraper, we devise a multiprocessing architecture while providing authentication<span> to the multiple processes based on the simulation of the user access behavior to Twitter. This allows us to maximize the parallelism of crawling even in a single machine. Through extensive experiments, we show that DeepScraper can crawl the entire tweets of 99 users, which amounts to 5,798,052 tweets while Twitter standard API can crawl only 243,650 tweets of them due to the constraints of the number of tweets to scrape. In other words, DeepScraper could collect 23.7 times more tweets for the 99 users than the standard API. We also show the efficiency of DeepScraper. First, we show the effect of the authenticated multiprocessing by showing that it increases the crawling speed from 2.03</span></span></span><span><math><mo>∼</mo></math></span>10.57 times as the number of running processes increases from 2 to 32 compared to DeepScraper with a single process. Then, we compare the crawling speed of DeepScraper with the existing studies. The result shows that DeepScraper is compared to even Twitter standard APIs and Twitter4J while DeepScraper can scrape much more tweets than them. Furthermore, DeepScraper is much faster than Twitter Scrapy roughly 3.69 times while both can scrape the entire tweets for the target users or keywords.</p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"149 ","pages":"Article 102260"},"PeriodicalIF":2.5,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138534226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"S_IDS: An efficient skyline query algorithm over incomplete data streams","authors":"Mei Bai,&nbsp;Yuxue Han,&nbsp;Peng Yin,&nbsp;Xite Wang,&nbsp;Guanyu Li,&nbsp;Bo Ning,&nbsp;Qian Ma","doi":"10.1016/j.datak.2023.102258","DOIUrl":"10.1016/j.datak.2023.102258","url":null,"abstract":"<div><p>The efficient processing of mass stream data has attracted wide attention in the database field. The skyline query on the sensor data stream can monitor multiple targets in real time, to avoid abnormal events such as fire and explosion, which is very useful in the practical application of sensor data monitoring. However, real-world stream data may often contain incomplete data attributes due to faulty sensing devices or imperfect data collection techniques. Skyline queries over incomplete data streams may lead to a lack of transitivity and loop domination issues. To solve the problem of the skyline query over incomplete data streams, firstly, this paper uses differential dependency rule (DD) to fill the missing attribute values of data in the incomplete data stream. Then, the hierarchical grid index (HGrid) is introduced into the field of skyline query to improve pruning efficiency. In the process of skyline calculation, this paper only keeps as few calculation results as possible for the data that may affect the result to avoid a large number of repeated calculations. Thus, S_IDS (Skyline query algorithm over Incomplete Data Stream) is proposed to query skyline results with high confidence from the incomplete data stream. Finally, by comparing with the most advanced skyline query algorithms over incomplete data streams, the correctness and efficiency of the proposed S_IDS algorithm are proved.</p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"149 ","pages":"Article 102258"},"PeriodicalIF":2.5,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138534229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Explainable influenza forecasting scheme using DCC-based feature selection","authors":"Sungwoo Park ,&nbsp;Jaeuk Moon ,&nbsp;Seungwon Jung ,&nbsp;Seungmin Rho ,&nbsp;Eenjun Hwang","doi":"10.1016/j.datak.2023.102256","DOIUrl":"https://doi.org/10.1016/j.datak.2023.102256","url":null,"abstract":"<div><p>As influenza is easily converted to another type of virus and spreads very quickly from person to person, it is more likely to develop into a pandemic. Even though vaccines are the most effective way to prevent influenza, it takes a lot of time to produce them. Due to this, there has been an imbalance in the supply and demand of influenza vaccines every year. For a smooth vaccine supply, it is necessary to accurately forecast vaccine demand at least three to six months in advance. So far, many machine learning-based predictive models have shown excellent performance. However, their use was limited due to performance deterioration due to inappropriate training data and inability to explain the results. To solve these problems, in this paper, we propose an explainable influenza forecasting model. In particular, the model selects highly related data based on the distance correlation coefficient for effective training and explains the prediction results using shapley additive explanations. We evaluated its performance through extensive experiments. We report some of the results.</p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"149 ","pages":"Article 102256"},"PeriodicalIF":2.5,"publicationDate":"2023-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138471983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A-MKMC: An effective adaptive-based multilevel K-means clustering with optimal centroid selection using hybrid heuristic approach for handling the incomplete data","authors":"Hima Vijayan ,&nbsp;Subramaniam M ,&nbsp;Sathiyasekar K","doi":"10.1016/j.datak.2023.102243","DOIUrl":"10.1016/j.datak.2023.102243","url":null,"abstract":"<div><p><span><span>In general, clustering is defined as partitioning similar and dissimilar objects into several groups. It has been widely used in applications like pattern recognition, image processing, and data analysis. When the dataset contains some missing data or value, it is termed incomplete data. In such implications, the incomplete dataset issue is untreatable while validating the data. Due to these flaws, the quality or standard level of the data gets an impact. Hence, the handling of missing values is done by influencing the clustering mechanisms for sorting out the missing data. Yet, the traditional </span>clustering algorithms<span> fail to combat the issues as it is not supposed to maintain large dimensional data. It is also caused by errors of human intervention or inaccurate outcomes. To alleviate the challenging issue of incomplete data, a novel clustering algorithm is proposed. Initially, incomplete or mixed data is garnered from the five different standard data sources. Once the data is to be collected, it is undergone the pre-processing phase, which is accomplished using data normalization. Subsequently, the final step is processed by the new clustering algorithm that is termed Adaptive centroid based Multilevel K-Means Clustering (A-MKMC), in which the cluster centroid is optimized by integrating the two conventional algorithms such as Border Collie Optimization (BCO) and </span></span>Whale Optimization Algorithm<span> (WOA) named as Hybrid Border Collie Whale Optimization (HBCWO). Therefore, the validation of the novel clustering model is estimated using various measures and compared against traditional mechanisms. From the overall result analysis, the accuracy and precision of the designed HBCWO-A-MKMC method attain 93 % and 95 %. Hence, the adaptive clustering process exploits the higher performance that aids in sorting out the missing data issuecompared to the other conventional methods.</span></p></div>","PeriodicalId":55184,"journal":{"name":"Data & Knowledge Engineering","volume":"150 ","pages":"Article 102243"},"PeriodicalIF":2.5,"publicationDate":"2023-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138534224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}