Pub Date : 2023-02-27DOI: 10.1108/dta-06-2021-0156
Vasileios Stamatis, M. Salampasis, K. Diamantaras
PurposeIn federated search, a query is sent simultaneously to multiple resources and each one of them returns a list of results. These lists are merged into a single list using the results merging process. In this work, the authors apply machine learning methods for results merging in federated patent search. Even though several methods for results merging have been developed, none of them were tested on patent data nor considered several machine learning models. Thus, the authors experiment with state-of-the-art methods using patent data and they propose two new methods for results merging that use machine learning models.Design/methodology/approachThe methods are based on a centralized index containing samples of documents from all the remote resources, and they implement machine learning models to estimate comparable scores for the documents retrieved by different resources. The authors examine the new methods in cooperative and uncooperative settings where document scores from the remote search engines are available and not, respectively. In uncooperative environments, they propose two methods for assigning document scores.FindingsThe effectiveness of the new results merging methods was measured against state-of-the-art models and found to be superior to them in many cases with significant improvements. The random forest model achieves the best results in comparison to all other models and presents new insights for the results merging problem.Originality/valueIn this article the authors prove that machine learning models can substitute other standard methods and models that used for results merging for many years. Our methods outperformed state-of-the-art estimation methods for results merging, and they proved that they are more effective for federated patent search.
{"title":"Machine learning methods for results merging in patent retrieval","authors":"Vasileios Stamatis, M. Salampasis, K. Diamantaras","doi":"10.1108/dta-06-2021-0156","DOIUrl":"https://doi.org/10.1108/dta-06-2021-0156","url":null,"abstract":"PurposeIn federated search, a query is sent simultaneously to multiple resources and each one of them returns a list of results. These lists are merged into a single list using the results merging process. In this work, the authors apply machine learning methods for results merging in federated patent search. Even though several methods for results merging have been developed, none of them were tested on patent data nor considered several machine learning models. Thus, the authors experiment with state-of-the-art methods using patent data and they propose two new methods for results merging that use machine learning models.Design/methodology/approachThe methods are based on a centralized index containing samples of documents from all the remote resources, and they implement machine learning models to estimate comparable scores for the documents retrieved by different resources. The authors examine the new methods in cooperative and uncooperative settings where document scores from the remote search engines are available and not, respectively. In uncooperative environments, they propose two methods for assigning document scores.FindingsThe effectiveness of the new results merging methods was measured against state-of-the-art models and found to be superior to them in many cases with significant improvements. The random forest model achieves the best results in comparison to all other models and presents new insights for the results merging problem.Originality/valueIn this article the authors prove that machine learning models can substitute other standard methods and models that used for results merging for many years. Our methods outperformed state-of-the-art estimation methods for results merging, and they proved that they are more effective for federated patent search.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44771332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-25DOI: 10.1108/dta-03-2022-0093
Bin Wang, Huifeng Li, Le Tong, Qian Zhang, Sulei Zhu, Tao Yang
PurposeThis paper aims to address the following issues: (1) most existing methods are based on recurrent network, which is time-consuming to train long sequences due to not allowing for full parallelism; (2) personalized preference generally are not considered reasonably; (3) existing methods rarely systematically studied how to efficiently utilize various auxiliary information (e.g. user ID and time stamp) in trajectory data and the spatiotemporal relations among nonconsecutive locations.Design/methodology/approachThe authors propose a novel self-attention network–based model named SanMove to predict the next location via capturing the long- and short-term mobility patterns of users. Specifically, SanMove uses a self-attention module to capture each user's long-term preference, which can represent her personalized location preference. Meanwhile, the authors use a spatial-temporal guided noninvasive self-attention (STNOVA) module to exploit auxiliary information in the trajectory data to learn the user's short-term preference.FindingsThe authors evaluate SanMove on two real-world datasets. The experimental results demonstrate that SanMove is not only faster than the state-of-the-art recurrent neural network (RNN) based predict model but also outperforms the baselines for next location prediction.Originality/valueThe authors propose a self-attention-based sequential model named SanMove to predict the user's trajectory, which comprised long-term and short-term preference learning modules. SanMove allows full parallel processing of trajectories to improve processing efficiency. They propose an STNOVA module to capture the sequential transitions of current trajectories. Moreover, the self-attention module is used to process historical trajectory sequences in order to capture the personalized location preference of each user. The authors conduct extensive experiments on two check-in datasets. The experimental results demonstrate that the model has a fast training speed and excellent performance compared with the existing RNN-based methods for next location prediction.
{"title":"SanMove: next location recommendation via self-attention network","authors":"Bin Wang, Huifeng Li, Le Tong, Qian Zhang, Sulei Zhu, Tao Yang","doi":"10.1108/dta-03-2022-0093","DOIUrl":"https://doi.org/10.1108/dta-03-2022-0093","url":null,"abstract":"PurposeThis paper aims to address the following issues: (1) most existing methods are based on recurrent network, which is time-consuming to train long sequences due to not allowing for full parallelism; (2) personalized preference generally are not considered reasonably; (3) existing methods rarely systematically studied how to efficiently utilize various auxiliary information (e.g. user ID and time stamp) in trajectory data and the spatiotemporal relations among nonconsecutive locations.Design/methodology/approachThe authors propose a novel self-attention network–based model named SanMove to predict the next location via capturing the long- and short-term mobility patterns of users. Specifically, SanMove uses a self-attention module to capture each user's long-term preference, which can represent her personalized location preference. Meanwhile, the authors use a spatial-temporal guided noninvasive self-attention (STNOVA) module to exploit auxiliary information in the trajectory data to learn the user's short-term preference.FindingsThe authors evaluate SanMove on two real-world datasets. The experimental results demonstrate that SanMove is not only faster than the state-of-the-art recurrent neural network (RNN) based predict model but also outperforms the baselines for next location prediction.Originality/valueThe authors propose a self-attention-based sequential model named SanMove to predict the user's trajectory, which comprised long-term and short-term preference learning modules. SanMove allows full parallel processing of trajectories to improve processing efficiency. They propose an STNOVA module to capture the sequential transitions of current trajectories. Moreover, the self-attention module is used to process historical trajectory sequences in order to capture the personalized location preference of each user. The authors conduct extensive experiments on two check-in datasets. The experimental results demonstrate that the model has a fast training speed and excellent performance compared with the existing RNN-based methods for next location prediction.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76459007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-25DOI: 10.1108/dta-03-2022-0107
A. Z. Abbasi, Natasha Ayaz, Sana Kanwal, M. Albashrawi, Nadine Khair
PurposeTikTok social media app has become one of the most popular forms of leisure and entertainment activities, but how hedonic consumption experiences (comprising fantasy, escapism, enjoyment, role projection, sensory, arousal and emotional involvement) of the TikTok app determine users' intention to use the app and its resulting impact on the actual usage behavior remains limited in the information systems literature, especially featuring the hedonic consumption perspective in entertainment industry.Design/methodology/approachThis study employs uses & gratification theory to answer the “why” via predicting the role of hedonic consumption experiences that serve as gratifications to trigger technology acceptance behavior (especially, in form of users' behavioral intention to use the TikTok app and its further impact on usage behavior). This study utilizes the partial least squares-structural equation modeling approach to perform data analyses on 258 TikTok app users.FindingsOur results provide a strong support such that users' playful consumption experiences (i.e. escapism, role projection, arousal, sensory experience and enjoyment) positively influence their intention to use the TikTok app and its resultant effect on users' actual usage of the app. In contrast, fantasy and emotional involvement fail to influence users' intention to use the TikTok app.Originality/valueTo the best of our knowledge, our investigation is one of the first studies to apply the hedonic consumption experiences as potential gratifications that derive users' intention and its subsequent influence on the actual usage of the TikTok app. Our study results would assist marketing and brand managers to redefine approaches and tactics to create effective strategies that implement essential determinants to increase behavioral intention among entertainment service providers.
{"title":"TikTok app usage behavior: the role of hedonic consumption experiences","authors":"A. Z. Abbasi, Natasha Ayaz, Sana Kanwal, M. Albashrawi, Nadine Khair","doi":"10.1108/dta-03-2022-0107","DOIUrl":"https://doi.org/10.1108/dta-03-2022-0107","url":null,"abstract":"PurposeTikTok social media app has become one of the most popular forms of leisure and entertainment activities, but how hedonic consumption experiences (comprising fantasy, escapism, enjoyment, role projection, sensory, arousal and emotional involvement) of the TikTok app determine users' intention to use the app and its resulting impact on the actual usage behavior remains limited in the information systems literature, especially featuring the hedonic consumption perspective in entertainment industry.Design/methodology/approachThis study employs uses & gratification theory to answer the “why” via predicting the role of hedonic consumption experiences that serve as gratifications to trigger technology acceptance behavior (especially, in form of users' behavioral intention to use the TikTok app and its further impact on usage behavior). This study utilizes the partial least squares-structural equation modeling approach to perform data analyses on 258 TikTok app users.FindingsOur results provide a strong support such that users' playful consumption experiences (i.e. escapism, role projection, arousal, sensory experience and enjoyment) positively influence their intention to use the TikTok app and its resultant effect on users' actual usage of the app. In contrast, fantasy and emotional involvement fail to influence users' intention to use the TikTok app.Originality/valueTo the best of our knowledge, our investigation is one of the first studies to apply the hedonic consumption experiences as potential gratifications that derive users' intention and its subsequent influence on the actual usage of the TikTok app. Our study results would assist marketing and brand managers to redefine approaches and tactics to create effective strategies that implement essential determinants to increase behavioral intention among entertainment service providers.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77615971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-25DOI: 10.1108/dta-10-2021-0276
Ningya Wang, Yang Zhao, Ruoxin Zhou
PurposeAs a derivative model of e-commerce, social commerce has received increasing attention in recent years. Empirical studies on social commerce have examined the key factors that influence users' attitudes or adoption intentions, but their conclusions are context-based and are not entirely consistent. This study aims to draw a general conclusion by systematically synthesizing the findings of previous studies and examine whether cultural differences play a moderating role in users' social commerce adoption.Design/methodology/approachA meta-analysis based on 11,786 independent samples from 39 empirical studies was conducted to integrate their results and develop a comprehensive conceptual model. A moderator analysis was carried out to investigate the moderating effect of culture by dividing the context into subgroups of individualistic and collectivistic cultures.FindingsThe results show that this comprehensive conceptual model can help better understand the adoption of social commerce. Meanwhile, the moderator analysis indicates that cultural differences have a significant moderating effect on the relationship between the determinants and the adoption of social commerce.Originality/valueThe findings of this paper have theoretical implications and make managerial contributions.
{"title":"A meta-analysis of social commerce adoption and the moderating effect of culture","authors":"Ningya Wang, Yang Zhao, Ruoxin Zhou","doi":"10.1108/dta-10-2021-0276","DOIUrl":"https://doi.org/10.1108/dta-10-2021-0276","url":null,"abstract":"PurposeAs a derivative model of e-commerce, social commerce has received increasing attention in recent years. Empirical studies on social commerce have examined the key factors that influence users' attitudes or adoption intentions, but their conclusions are context-based and are not entirely consistent. This study aims to draw a general conclusion by systematically synthesizing the findings of previous studies and examine whether cultural differences play a moderating role in users' social commerce adoption.Design/methodology/approachA meta-analysis based on 11,786 independent samples from 39 empirical studies was conducted to integrate their results and develop a comprehensive conceptual model. A moderator analysis was carried out to investigate the moderating effect of culture by dividing the context into subgroups of individualistic and collectivistic cultures.FindingsThe results show that this comprehensive conceptual model can help better understand the adoption of social commerce. Meanwhile, the moderator analysis indicates that cultural differences have a significant moderating effect on the relationship between the determinants and the adoption of social commerce.Originality/valueThe findings of this paper have theoretical implications and make managerial contributions.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"62066178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-24DOI: 10.1108/dta-05-2022-0215
Hyogon Kim, Eunmi Lee, Donghee Yoo
PurposeThis study quantified companies' views on the COVID-19 pandemic with sentiment analysis of US public companies' disclosures. The study aims to provide timely insights to shareholders, investors and consumers by exploring sentiment trends and changes in the industry and the relationship with stock price indices.Design/methodology/approachFrom more than 50,000 Form 10-K and Form 10-Q published between 2020 and 2021, over one million texts related to the COVID-19 pandemic were extracted. Applying the FinBERT fine-tuned for this study, the texts were classified into positive, negative and neutral sentiments. The correlations between sentiment trends, differences in sentiment distribution by industry and stock price indices were investigated by statistically testing the changes and distribution of quantified sentiments.FindingsFirst, there were quantitative changes in texts related to the COVID-19 pandemic in the US companies' disclosures. In addition, the changes in the trend of positive and negative sentiments were found. Second, industry patterns of positive and negative sentiment changes were similar, but no similarities were found in neutral sentiments. Third, in analyzing the relationship between the representative US stock indices and the sentiment trends, the results indicated a positive relationship with positive sentiments and a negative relationship with negative sentiments.Originality/valuePerforming sentiment analysis on formal documents like Securities and Exchange Commission (SEC) filings, this study was differentiated from previous studies by revealing the quantitative changes of sentiment implied in the documents and the trend over time. Moreover, an appropriate data preprocessing procedure and analysis method were presented for the time-series analysis of the SEC filings.
{"title":"Do SEC filings indicate any trends? Evidence from the sentiment distribution of forms 10-K and 10-Q with FinBERT","authors":"Hyogon Kim, Eunmi Lee, Donghee Yoo","doi":"10.1108/dta-05-2022-0215","DOIUrl":"https://doi.org/10.1108/dta-05-2022-0215","url":null,"abstract":"PurposeThis study quantified companies' views on the COVID-19 pandemic with sentiment analysis of US public companies' disclosures. The study aims to provide timely insights to shareholders, investors and consumers by exploring sentiment trends and changes in the industry and the relationship with stock price indices.Design/methodology/approachFrom more than 50,000 Form 10-K and Form 10-Q published between 2020 and 2021, over one million texts related to the COVID-19 pandemic were extracted. Applying the FinBERT fine-tuned for this study, the texts were classified into positive, negative and neutral sentiments. The correlations between sentiment trends, differences in sentiment distribution by industry and stock price indices were investigated by statistically testing the changes and distribution of quantified sentiments.FindingsFirst, there were quantitative changes in texts related to the COVID-19 pandemic in the US companies' disclosures. In addition, the changes in the trend of positive and negative sentiments were found. Second, industry patterns of positive and negative sentiment changes were similar, but no similarities were found in neutral sentiments. Third, in analyzing the relationship between the representative US stock indices and the sentiment trends, the results indicated a positive relationship with positive sentiments and a negative relationship with negative sentiments.Originality/valuePerforming sentiment analysis on formal documents like Securities and Exchange Commission (SEC) filings, this study was differentiated from previous studies by revealing the quantitative changes of sentiment implied in the documents and the trend over time. Moreover, an appropriate data preprocessing procedure and analysis method were presented for the time-series analysis of the SEC filings.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81593753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-24DOI: 10.1108/dta-08-2022-0330
F. Nakach, Hasnae Zerouaoui, A. Idri
PurposeHistopathology biopsy imaging is currently the gold standard for the diagnosis of breast cancer in clinical practice. Pathologists examine the images at various magnifications to identify the type of tumor because if only one magnification is taken into account, the decision may not be accurate. This study explores the performance of transfer learning and late fusion to construct multi-scale ensembles that fuse different magnification-specific deep learning models for the binary classification of breast tumor slides.Design/methodology/approachThree pretrained deep learning techniques (DenseNet 201, MobileNet v2 and Inception v3) were used to classify breast tumor images over the four magnification factors of the Breast Cancer Histopathological Image Classification dataset (40×, 100×, 200× and 400×). To fuse the predictions of the models trained on different magnification factors, different aggregators were used, including weighted voting and seven meta-classifiers trained on slide predictions using class labels and the probabilities assigned to each class. The best cluster of the outperforming models was chosen using the Scott–Knott statistical test, and the top models were ranked using the Borda count voting system.FindingsThis study recommends the use of transfer learning and late fusion for histopathological breast cancer image classification by constructing multi-magnification ensembles because they perform better than models trained on each magnification separately.Originality/valueThe best multi-scale ensembles outperformed state-of-the-art integrated models and achieved an accuracy mean value of 98.82 per cent, precision of 98.46 per cent, recall of 100 per cent and F1-score of 99.20 per cent.
{"title":"Binary classification of multi-magnification histopathological breast cancer images using late fusion and transfer learning","authors":"F. Nakach, Hasnae Zerouaoui, A. Idri","doi":"10.1108/dta-08-2022-0330","DOIUrl":"https://doi.org/10.1108/dta-08-2022-0330","url":null,"abstract":"PurposeHistopathology biopsy imaging is currently the gold standard for the diagnosis of breast cancer in clinical practice. Pathologists examine the images at various magnifications to identify the type of tumor because if only one magnification is taken into account, the decision may not be accurate. This study explores the performance of transfer learning and late fusion to construct multi-scale ensembles that fuse different magnification-specific deep learning models for the binary classification of breast tumor slides.Design/methodology/approachThree pretrained deep learning techniques (DenseNet 201, MobileNet v2 and Inception v3) were used to classify breast tumor images over the four magnification factors of the Breast Cancer Histopathological Image Classification dataset (40×, 100×, 200× and 400×). To fuse the predictions of the models trained on different magnification factors, different aggregators were used, including weighted voting and seven meta-classifiers trained on slide predictions using class labels and the probabilities assigned to each class. The best cluster of the outperforming models was chosen using the Scott–Knott statistical test, and the top models were ranked using the Borda count voting system.FindingsThis study recommends the use of transfer learning and late fusion for histopathological breast cancer image classification by constructing multi-magnification ensembles because they perform better than models trained on each magnification separately.Originality/valueThe best multi-scale ensembles outperformed state-of-the-art integrated models and achieved an accuracy mean value of 98.82 per cent, precision of 98.46 per cent, recall of 100 per cent and F1-score of 99.20 per cent.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45716506","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-22DOI: 10.1108/dta-10-2022-0406
Meriem Laifa, Djamila Mohdeb
PurposeThis study provides an overview of the application of sentiment analysis (SA) in exploring social movements (SMs). It also compares different models for a SA task of Algerian Arabic tweets related to early days of the Algerian SM, called Hirak.Design/methodology/approachRelated tweets were retrieved using relevant hashtags followed by multiple data cleaning procedures. Foundational machine learning methods such as Naive Bayes, Support Vector Machine, Logistic Regression (LR) and Decision Tree were implemented. For each classifier, two feature extraction techniques were used and compared, namely Bag of Words and Term Frequency–Inverse Document Frequency. Moreover, three fine-tuned pretrained transformers AraBERT and DziriBERT and the multilingual transformer XLM-R were used for the comparison.FindingsThe findings of this paper emphasize the vital role social media played during the Hirak. Results revealed that most individuals had a positive attitude toward the Hirak. Moreover, the presented experiments provided important insights into the possible use of both basic machine learning and transfer learning models to analyze SA of Algerian text datasets. When comparing machine learning models with transformers in terms of accuracy, precision, recall and F1-score, the results are fairly similar, with LR outperforming all models with a 68 per cent accuracy rate.Originality/valueAt the time of writing, the Algerian SM was not thoroughly investigated or discussed in the Computer Science literature. This analysis makes a limited but unique contribution to understanding the Algerian Hirak using artificial intelligence. This study proposes what it considers to be a unique basis for comprehending this event with the goal of generating a foundation for future studies by comparing different SA techniques on a low-resource language.
{"title":"Sentiment analysis of the Algerian social movement inception","authors":"Meriem Laifa, Djamila Mohdeb","doi":"10.1108/dta-10-2022-0406","DOIUrl":"https://doi.org/10.1108/dta-10-2022-0406","url":null,"abstract":"PurposeThis study provides an overview of the application of sentiment analysis (SA) in exploring social movements (SMs). It also compares different models for a SA task of Algerian Arabic tweets related to early days of the Algerian SM, called Hirak.Design/methodology/approachRelated tweets were retrieved using relevant hashtags followed by multiple data cleaning procedures. Foundational machine learning methods such as Naive Bayes, Support Vector Machine, Logistic Regression (LR) and Decision Tree were implemented. For each classifier, two feature extraction techniques were used and compared, namely Bag of Words and Term Frequency–Inverse Document Frequency. Moreover, three fine-tuned pretrained transformers AraBERT and DziriBERT and the multilingual transformer XLM-R were used for the comparison.FindingsThe findings of this paper emphasize the vital role social media played during the Hirak. Results revealed that most individuals had a positive attitude toward the Hirak. Moreover, the presented experiments provided important insights into the possible use of both basic machine learning and transfer learning models to analyze SA of Algerian text datasets. When comparing machine learning models with transformers in terms of accuracy, precision, recall and F1-score, the results are fairly similar, with LR outperforming all models with a 68 per cent accuracy rate.Originality/valueAt the time of writing, the Algerian SM was not thoroughly investigated or discussed in the Computer Science literature. This analysis makes a limited but unique contribution to understanding the Algerian Hirak using artificial intelligence. This study proposes what it considers to be a unique basis for comprehending this event with the goal of generating a foundation for future studies by comparing different SA techniques on a low-resource language.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46130946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
PurposeIntent detection and slot filling are two important tasks in question comprehension of a question answering system. This study aims to build a joint task model with some generalization ability and benchmark its performance over other neural network models mentioned in this paper.Design/methodology/approachThis study used a deep-learning-based approach for the joint modeling of question intent detection and slot filling. Meanwhile, the internal cell structure of the long short-term memory (LSTM) network was improved. Furthermore, the dataset Computer Science Literature Question (CSLQ) was constructed based on the Science and Technology Knowledge Graph. The datasets Airline Travel Information Systems, Snips (a natural language processing dataset of the consumer intent engine collected by Snips) and CSLQ were used for the empirical analysis. The accuracy of intent detection and F1 score of slot filling, as well as the semantic accuracy of sentences, were compared for several models.FindingsThe results showed that the proposed model outperformed all other benchmark methods, especially for the CSLQ dataset. This proves that the design of this study improved the comprehensive performance and generalization ability of the model to some extent.Originality/valueThis study contributes to the understanding of question sentences in a specific domain. LSTM was improved, and a computer literature domain dataset was constructed herein. This will lay the data and model foundation for the future construction of a computer literature question answering system.
目的意图检测和空位填充是问答系统问题理解中的两项重要任务。本研究旨在建立一个具有一定泛化能力的联合任务模型，并将其性能与本文中提到的其他神经网络模型进行比较。设计/方法论/方法本研究使用了一种基于深度学习的方法来对问题意图检测和空位填充进行联合建模。同时，长短期记忆（LSTM）网络的内部细胞结构得到了改善。此外，基于科学技术知识图谱构建了计算机科学文献问题数据集（CSLQ）。数据集Airline Travel Information Systems、Snipps（由Snipps收集的消费者意图引擎的自然语言处理数据集）和CSLQ用于实证分析。比较了几种模型的意图检测的准确性、空位填充的F1分数以及句子的语义准确性。结果表明，所提出的模型优于所有其他基准方法，尤其是对于CSLQ数据集。这证明了本研究的设计在一定程度上提高了模型的综合性能和泛化能力。独创性/价值这项研究有助于理解特定领域的疑问句。对LSTM进行了改进，构建了计算机文献领域数据集。这将为未来构建计算机文献问答系统奠定数据和模型基础。
{"title":"Joint modeling method of question intent detection and slot filling for domain-oriented question answering system","authors":"Huiyong Wang, Ding Yang, Liang Guo, Xiaoming Zhang","doi":"10.1108/dta-07-2022-0281","DOIUrl":"https://doi.org/10.1108/dta-07-2022-0281","url":null,"abstract":"PurposeIntent detection and slot filling are two important tasks in question comprehension of a question answering system. This study aims to build a joint task model with some generalization ability and benchmark its performance over other neural network models mentioned in this paper.Design/methodology/approachThis study used a deep-learning-based approach for the joint modeling of question intent detection and slot filling. Meanwhile, the internal cell structure of the long short-term memory (LSTM) network was improved. Furthermore, the dataset Computer Science Literature Question (CSLQ) was constructed based on the Science and Technology Knowledge Graph. The datasets Airline Travel Information Systems, Snips (a natural language processing dataset of the consumer intent engine collected by Snips) and CSLQ were used for the empirical analysis. The accuracy of intent detection and F1 score of slot filling, as well as the semantic accuracy of sentences, were compared for several models.FindingsThe results showed that the proposed model outperformed all other benchmark methods, especially for the CSLQ dataset. This proves that the design of this study improved the comprehensive performance and generalization ability of the model to some extent.Originality/valueThis study contributes to the understanding of question sentences in a specific domain. LSTM was improved, and a computer literature domain dataset was constructed herein. This will lay the data and model foundation for the future construction of a computer literature question answering system.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45614283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-06DOI: 10.1108/dta-02-2022-0056
Riju Bhattacharya, N. K. Nagwani, Sarsij Tripathi
PurposeA community demonstrates the unique qualities and relationships between its members that distinguish it from other communities within a network. Network analysis relies heavily on community detection. Despite the traditional spectral clustering and statistical inference methods, deep learning techniques for community detection have grown in popularity due to their ease of processing high-dimensional network data. Graph convolutional neural networks (GCNNs) have received much attention recently and have developed into a potential and ubiquitous method for directly detecting communities on graphs. Inspired by the promising results of graph convolutional networks (GCNs) in analyzing graph structure data, a novel community graph convolutional network (CommunityGCN) as a semi-supervised node classification model has been proposed and compared with recent baseline methods graph attention network (GAT), GCN-based technique for unsupervised community detection and Markov random fields combined with graph convolutional network (MRFasGCN).Design/methodology/approachThis work presents the method for identifying communities that combines the notion of node classification via message passing with the architecture of a semi-supervised graph neural network. Six benchmark datasets, namely, Cora, CiteSeer, ACM, Karate, IMDB and Facebook, have been used in the experimentation.FindingsIn the first set of experiments, the scaled normalized average matrix of all neighbor's features including the node itself was obtained, followed by obtaining the weighted average matrix of low-dimensional nodes. In the second set of experiments, the average weighted matrix was forwarded to the GCN with two layers and the activation function for predicting the node class was applied. The results demonstrate that node classification with GCN can improve the performance of identifying communities on graph datasets.Originality/valueThe experiment reveals that the CommunityGCN approach has given better results with accuracy, normalized mutual information, F1 and modularity scores of 91.26, 79.9, 92.58 and 70.5 per cent, respectively, for detecting communities in the graph network, which is much greater than the range of 55.7–87.07 per cent reported in previous literature. Thus, it has been concluded that the GCN with node classification models has improved the accuracy.
{"title":"CommunityGCN: community detection using node classification with graph convolution network","authors":"Riju Bhattacharya, N. K. Nagwani, Sarsij Tripathi","doi":"10.1108/dta-02-2022-0056","DOIUrl":"https://doi.org/10.1108/dta-02-2022-0056","url":null,"abstract":"PurposeA community demonstrates the unique qualities and relationships between its members that distinguish it from other communities within a network. Network analysis relies heavily on community detection. Despite the traditional spectral clustering and statistical inference methods, deep learning techniques for community detection have grown in popularity due to their ease of processing high-dimensional network data. Graph convolutional neural networks (GCNNs) have received much attention recently and have developed into a potential and ubiquitous method for directly detecting communities on graphs. Inspired by the promising results of graph convolutional networks (GCNs) in analyzing graph structure data, a novel community graph convolutional network (CommunityGCN) as a semi-supervised node classification model has been proposed and compared with recent baseline methods graph attention network (GAT), GCN-based technique for unsupervised community detection and Markov random fields combined with graph convolutional network (MRFasGCN).Design/methodology/approachThis work presents the method for identifying communities that combines the notion of node classification via message passing with the architecture of a semi-supervised graph neural network. Six benchmark datasets, namely, Cora, CiteSeer, ACM, Karate, IMDB and Facebook, have been used in the experimentation.FindingsIn the first set of experiments, the scaled normalized average matrix of all neighbor's features including the node itself was obtained, followed by obtaining the weighted average matrix of low-dimensional nodes. In the second set of experiments, the average weighted matrix was forwarded to the GCN with two layers and the activation function for predicting the node class was applied. The results demonstrate that node classification with GCN can improve the performance of identifying communities on graph datasets.Originality/valueThe experiment reveals that the CommunityGCN approach has given better results with accuracy, normalized mutual information, F1 and modularity scores of 91.26, 79.9, 92.58 and 70.5 per cent, respectively, for detecting communities in the graph network, which is much greater than the range of 55.7–87.07 per cent reported in previous literature. Thus, it has been concluded that the GCN with node classification models has improved the accuracy.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41895655","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-02-03DOI: 10.1108/dta-05-2022-0199
Lizhao Zhang, Jui-Long Hung, Xu Du, Hao Li, Zhuang Hu
PurposeStudent engagement is a key factor that connects with student achievement and retention. This paper aims to identify individuals' engagement automatically in the classroom with multimodal data for supporting educational research.Design/methodology/approachThe video and electroencephalogram data of 36 undergraduates were collected to represent observable and internal information. Since different modal data have different granularity, this study proposed the Fast–Slow Neural Network (FSNN) to detect engagement through both observable and internal information, with an asynchrony structure to preserve the sequence information of data with different granularity.FindingsExperimental results show that the proposed algorithm can recognize engagement better than the traditional data fusion methods. The results are also analyzed to figure out the reasons for the better performance of the proposed FSNN.Originality/valueThis study combined multimodal data from observable and internal aspects to improve the accuracy of engagement detection in the classroom. The proposed FSNN used the asynchronous process to deal with the problem of remaining sequential information when facing multimodal data with different granularity.
{"title":"Multimodal Fast-Slow Neural Network for learning engagement evaluation","authors":"Lizhao Zhang, Jui-Long Hung, Xu Du, Hao Li, Zhuang Hu","doi":"10.1108/dta-05-2022-0199","DOIUrl":"https://doi.org/10.1108/dta-05-2022-0199","url":null,"abstract":"PurposeStudent engagement is a key factor that connects with student achievement and retention. This paper aims to identify individuals' engagement automatically in the classroom with multimodal data for supporting educational research.Design/methodology/approachThe video and electroencephalogram data of 36 undergraduates were collected to represent observable and internal information. Since different modal data have different granularity, this study proposed the Fast–Slow Neural Network (FSNN) to detect engagement through both observable and internal information, with an asynchrony structure to preserve the sequence information of data with different granularity.FindingsExperimental results show that the proposed algorithm can recognize engagement better than the traditional data fusion methods. The results are also analyzed to figure out the reasons for the better performance of the proposed FSNN.Originality/valueThis study combined multimodal data from observable and internal aspects to improve the accuracy of engagement detection in the classroom. The proposed FSNN used the asynchronous process to deal with the problem of remaining sequential information when facing multimodal data with different granularity.","PeriodicalId":56156,"journal":{"name":"Data Technologies and Applications","volume":null,"pages":null},"PeriodicalIF":1.6,"publicationDate":"2023-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77554130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","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：