Question routing (QR) aims to route newly submitted questions to the potential experts most likely to provide answers. Many previous works formalize the question routing task as a text matching and ranking problem between questions and user profiles, focusing on text representation and semantic similarity computation. However, these works often fail to extract matching features efficiently and lack deep contextual textual understanding. Moreover, we argue that in addition to the semantic similarity between terms, the interactive relationship between question sequences and user profile sequences also plays an important role in matching. In this paper, we proposed two BERT-based models called QR-BERTrep and QR-tBERTint to address these issues from different perspectives. QR-BERTrep is a representation-based feature ensemble model in which we integrated a weighted sum of BERT layer outputs as an extra feature into a Siamese deep matching network, aiming to address the non-context-aware word embedding and limited semantic understanding. QR-tBERTint is an interaction-based model that explores the interactive relationships between sequences as well as the semantic similarity of terms through a topic-enhanced BERT model. Specifically, it fuses a short-text-friendly topic model to capture corpus-level semantic information. Experimental results on real-world data demonstrate that QR-BERTrep significantly outperforms other traditional representation-based models. Meanwhile, QR-tBERTint exceeds QR-BERTrep and QR-BERTint with a maximum increase of 17.26% and 11.52% in MAP, respectively, showing that combining global topic information and exploring interactive relationships between sequences is quite effective for question routing tasks.
{"title":"Deep Semantic Understanding and Sequence Relevance Learning for Question Routing in Community Question Answering","authors":"Hong Li, Jianjun Li, Guohui Li, Chunzhi Wang, Wenjun Cao, Zixuan Chen","doi":"10.5755/j01.itc.52.3.33449","DOIUrl":"https://doi.org/10.5755/j01.itc.52.3.33449","url":null,"abstract":"Question routing (QR) aims to route newly submitted questions to the potential experts most likely to provide answers. Many previous works formalize the question routing task as a text matching and ranking problem between questions and user profiles, focusing on text representation and semantic similarity computation. However, these works often fail to extract matching features efficiently and lack deep contextual textual understanding. Moreover, we argue that in addition to the semantic similarity between terms, the interactive relationship between question sequences and user profile sequences also plays an important role in matching. In this paper, we proposed two BERT-based models called QR-BERTrep and QR-tBERTint to address these issues from different perspectives. QR-BERTrep is a representation-based feature ensemble model in which we integrated a weighted sum of BERT layer outputs as an extra feature into a Siamese deep matching network, aiming to address the non-context-aware word embedding and limited semantic understanding. QR-tBERTint is an interaction-based model that explores the interactive relationships between sequences as well as the semantic similarity of terms through a topic-enhanced BERT model. Specifically, it fuses a short-text-friendly topic model to capture corpus-level semantic information. Experimental results on real-world data demonstrate that QR-BERTrep significantly outperforms other traditional representation-based models. Meanwhile, QR-tBERTint exceeds QR-BERTrep and QR-BERTint with a maximum increase of 17.26% and 11.52% in MAP, respectively, showing that combining global topic information and exploring interactive relationships between sequences is quite effective for question routing tasks.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"2012 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134886471","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 : 2023-09-26DOI: 10.5755/j01.itc.52.3.31945
K. Ezhilarasi, D. Mansoor Hussain, M. Sowmiya, N. Krishnamoorthy
Currently, on the Internet, the information about agriculture is augmenting extremely; thus, searching for precise, relevant data of various details is highly complicated. To deal with particular difficulties like lower relevancy rate, false detection of retrieval resources, poor similarity rate, unstructured data format, multivariate data, irrelevant spelling, and higher computation time, an intelligent Information Retrieval (IR) system is required. An IR Framework centered on Levenshtein Distance Weight-centric Ontology (LDW-Ontology) and Sutskever Nesterov Momentum-centred Bidirectional Encoder Representation from Transformer (SNM-BERT) methodologies is presented here to overcome the complications as mentioned earlier. Firstly, the data is pre-processed, transmuting the unstructured data into a structured format, thus mitigating the error probabilities. Then, the LDW-Crop Ontology construction is done regarding the structured data. In the methodology presented, significance, frequency,and the suggestion of word in mind are considered to build Crop ontology. In the MongoDB database, the data being constructed are amassed. Then, by utilizing SNM-BERT, the data is trained for IR regarding clustered input produced by Inter Quartile Pruning Range-centred Hierarchical Divisive Clustering (IQPR-HDC) model. The LDW is computed for the provided user query; subsequently, the similarity evaluation outcomes are obtained from the database. The experiential evaluation displays that when analogized with the prevailing methodologies, a better accuracy of 94 % for simple queries and 92% for complex queries is achieved. Along with retrieval rate with lower computation time is achieved by the proposed methodology.
{"title":"Crop Information Retrieval Framework Based on LDW-Ontology and SNM-BERT Techniques","authors":"K. Ezhilarasi, D. Mansoor Hussain, M. Sowmiya, N. Krishnamoorthy","doi":"10.5755/j01.itc.52.3.31945","DOIUrl":"https://doi.org/10.5755/j01.itc.52.3.31945","url":null,"abstract":"Currently, on the Internet, the information about agriculture is augmenting extremely; thus, searching for precise, relevant data of various details is highly complicated. To deal with particular difficulties like lower relevancy rate, false detection of retrieval resources, poor similarity rate, unstructured data format, multivariate data, irrelevant spelling, and higher computation time, an intelligent Information Retrieval (IR) system is required. An IR Framework centered on Levenshtein Distance Weight-centric Ontology (LDW-Ontology) and Sutskever Nesterov Momentum-centred Bidirectional Encoder Representation from Transformer (SNM-BERT) methodologies is presented here to overcome the complications as mentioned earlier. Firstly, the data is pre-processed, transmuting the unstructured data into a structured format, thus mitigating the error probabilities. Then, the LDW-Crop Ontology construction is done regarding the structured data. In the methodology presented, significance, frequency,and the suggestion of word in mind are considered to build Crop ontology. In the MongoDB database, the data being constructed are amassed. Then, by utilizing SNM-BERT, the data is trained for IR regarding clustered input produced by Inter Quartile Pruning Range-centred Hierarchical Divisive Clustering (IQPR-HDC) model. The LDW is computed for the provided user query; subsequently, the similarity evaluation outcomes are obtained from the database. The experiential evaluation displays that when analogized with the prevailing methodologies, a better accuracy of 94 % for simple queries and 92% for complex queries is achieved. Along with retrieval rate with lower computation time is achieved by the proposed methodology.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134886683","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 : 2023-09-26DOI: 10.5755/j01.itc.52.3.33092
Kangyong Yin, Haosheng Huang, Wei Liang, Hongwu Xiao, Lei Wang
Achieving efficient end-to-end file transfer is challenging in a narrow-band communication environment with high latency and high packet loss rate. The traditional TCP-based scheme and the UDP-based automatic retransmission scheme have defects in the transmission performance, which cannot meet the increasing user demands. This paper proposes a high-efficiency file transfer scheme based on random linear network coding and the Kalman filtering algorithm to implement efficient end-to-end file transfer in narrow-band environment. The scheme predicts the link quality of file transmission through the Kalman filter algorithm and designs an adaptive coding strategy for file transfer through random linear network coding. Experimental results show that the proposed method outperforms traditional file transfer schemes.
{"title":"Network Coding for Efficient File Transfer in Narrowband Environments","authors":"Kangyong Yin, Haosheng Huang, Wei Liang, Hongwu Xiao, Lei Wang","doi":"10.5755/j01.itc.52.3.33092","DOIUrl":"https://doi.org/10.5755/j01.itc.52.3.33092","url":null,"abstract":"Achieving efficient end-to-end file transfer is challenging in a narrow-band communication environment with high latency and high packet loss rate. The traditional TCP-based scheme and the UDP-based automatic retransmission scheme have defects in the transmission performance, which cannot meet the increasing user demands. This paper proposes a high-efficiency file transfer scheme based on random linear network coding and the Kalman filtering algorithm to implement efficient end-to-end file transfer in narrow-band environment. The scheme predicts the link quality of file transmission through the Kalman filter algorithm and designs an adaptive coding strategy for file transfer through random linear network coding. Experimental results show that the proposed method outperforms traditional file transfer schemes.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134886912","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 : 2023-09-26DOI: 10.5755/j01.itc.52.3.33659
N. Sasikumar, M. Senthilkumar
The diagnosis of early-stage lung cancer can be challenging due to several factors. Firstly, the asymptomatic nature of the disease means that it may not present any noticeable symptoms until it has progressed to later stages. Additionally, the use of computed tomography, which can be expensive and involves repetitive radiation exposure, can further complicate the diagnostic process. Even specialists may encounter difficulties when examining lung CT imagery to identify pulmonary nodules, particularly in the case of cell lung adenocarcinoma lesions.This paper suggests a unique deep learning-based Deep Convolutional Generative Adversarial Networks (DCGAN) model for lung cancer classification. The dataset utilized for the experimental purpose is accessed from the LUNA16 challenge database. This comprises 888 CT scans of the lungs. These images are initially segmented using Quick-CapsNet (QCN) model and applied with Red Deer Optimization (RDO) algorithm to extract the optimized features. Furthermore, the categorization between benign and malignant tumors is carried out using the DC-GAN model. The pulmonary nodule detection accuracy of the proposed model is 98.65%, indicating early-stage lung cancer. It is discovered to be superior to other existing techniques, such as sophisticated deep learning, straightforward machine learning, and hybrid methods applied to lung CT scans for nodule diagnosis. According to experimental findings, the suggested way can significantly help radiologists spot early lung cancer and facilitate prompt patient management.
{"title":"Deep Convolutional Generative Adversarial Networks for Automated Segmentation and Detection of Lung Adenocarcinoma Using Red Deer Optimization Algorithm","authors":"N. Sasikumar, M. Senthilkumar","doi":"10.5755/j01.itc.52.3.33659","DOIUrl":"https://doi.org/10.5755/j01.itc.52.3.33659","url":null,"abstract":"The diagnosis of early-stage lung cancer can be challenging due to several factors. Firstly, the asymptomatic nature of the disease means that it may not present any noticeable symptoms until it has progressed to later stages. Additionally, the use of computed tomography, which can be expensive and involves repetitive radiation exposure, can further complicate the diagnostic process. Even specialists may encounter difficulties when examining lung CT imagery to identify pulmonary nodules, particularly in the case of cell lung adenocarcinoma lesions.This paper suggests a unique deep learning-based Deep Convolutional Generative Adversarial Networks (DCGAN) model for lung cancer classification. The dataset utilized for the experimental purpose is accessed from the LUNA16 challenge database. This comprises 888 CT scans of the lungs. These images are initially segmented using Quick-CapsNet (QCN) model and applied with Red Deer Optimization (RDO) algorithm to extract the optimized features. Furthermore, the categorization between benign and malignant tumors is carried out using the DC-GAN model. The pulmonary nodule detection accuracy of the proposed model is 98.65%, indicating early-stage lung cancer. It is discovered to be superior to other existing techniques, such as sophisticated deep learning, straightforward machine learning, and hybrid methods applied to lung CT scans for nodule diagnosis. According to experimental findings, the suggested way can significantly help radiologists spot early lung cancer and facilitate prompt patient management.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134960683","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}
For the traditional A* algorithm has problems such as long paths, large number of nodes, and the demand for dynamic obstacle cannot be avoided in complex environment. A mobile robot dynamic path avoidance method will be improved to improve the A * algorithm and improve DWA algorithm Two map environments are used for simulation verification. First, the evaluation function and key node selection strategy are optimized for the A* algorithm, and redundant nodes are deleted; then the dynamic obstacle distance evaluation function is added to the DWA algorithm which for the purpose of the obstacle avoidance performance can be enhanced. The results about the improved A* algorithm reduces 12.20% and 58.33% in path length and number of turning points respectively compared with the traditional A* algorithm can be obviously grasped by the simulation experiment; by using the fusion algorithm whose purpose of using arcs instead of the straight lines is to turn more smoothly, and can be closest to the global optimum while avoiding dynamic obstacles to complete the search.
{"title":"Robot Path Planning Research Incorporating Improved A* Algorithm and DWA Algorithm","authors":"Shiya Qu, Guang Feng, Yuhang Jiang, Chunyu Han, Dingyuan Hu, Hongbin Liang","doi":"10.5755/j01.itc.52.3.32791","DOIUrl":"https://doi.org/10.5755/j01.itc.52.3.32791","url":null,"abstract":"For the traditional A* algorithm has problems such as long paths, large number of nodes, and the demand for dynamic obstacle cannot be avoided in complex environment. A mobile robot dynamic path avoidance method will be improved to improve the A * algorithm and improve DWA algorithm Two map environments are used for simulation verification. First, the evaluation function and key node selection strategy are optimized for the A* algorithm, and redundant nodes are deleted; then the dynamic obstacle distance evaluation function is added to the DWA algorithm which for the purpose of the obstacle avoidance performance can be enhanced. The results about the improved A* algorithm reduces 12.20% and 58.33% in path length and number of turning points respectively compared with the traditional A* algorithm can be obviously grasped by the simulation experiment; by using the fusion algorithm whose purpose of using arcs instead of the straight lines is to turn more smoothly, and can be closest to the global optimum while avoiding dynamic obstacles to complete the search.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134961159","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 : 2023-07-15DOI: 10.5755/j01.itc.52.2.33415
Juan Zhu, Yuqing Ma, Jipeng Huang, Lianming Wang
Image segmentation is one of the key steps of target recognition. In order to improve the accuracy of image segmentation, an image segmentation algorithm combining Pulse Coupled Neural Network(PCNN) and adaptive Glowworm Algorithm(GA) is proposed. The algorithm retains the advantages of the GA. Introduce the adaptive moving step size and the population optimal value as adjustment factors. Enhance the ability to solve the global optimal value, and takes the weighted sum of the cross entropy, information entropy and compactness of the image as the fitness function of the GA. Maintain the diversity of image features and improving the accuracy of image segmentation. Experimental results show that compared with other algorithms, the segmented image obtained by this algorithm has better visual effect and the segmentation performance has the best comprehensive performance. For the seven gray-scale images in the Berkeley segmentation dataset, the segmentation effect is improved by 10.85% compared with TDE algorithm, 9.22% compared with GA algorithm, and 22.58% compared with AUTO algorithm.
{"title":"Image Segmentation Combining Pulse Coupled Neural Network and Adaptive Glowworm Algorithm","authors":"Juan Zhu, Yuqing Ma, Jipeng Huang, Lianming Wang","doi":"10.5755/j01.itc.52.2.33415","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.33415","url":null,"abstract":"Image segmentation is one of the key steps of target recognition. In order to improve the accuracy of image segmentation, an image segmentation algorithm combining Pulse Coupled Neural Network(PCNN) and adaptive Glowworm Algorithm(GA) is proposed. The algorithm retains the advantages of the GA. Introduce the adaptive moving step size and the population optimal value as adjustment factors. Enhance the ability to solve the global optimal value, and takes the weighted sum of the cross entropy, information entropy and compactness of the image as the fitness function of the GA. Maintain the diversity of image features and improving the accuracy of image segmentation. Experimental results show that compared with other algorithms, the segmented image obtained by this algorithm has better visual effect and the segmentation performance has the best comprehensive performance. For the seven gray-scale images in the Berkeley segmentation dataset, the segmentation effect is improved by 10.85% compared with TDE algorithm, 9.22% compared with GA algorithm, and 22.58% compared with AUTO algorithm.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"20 1","pages":"487-499"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88025873","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 : 2023-07-15DOI: 10.5755/j01.itc.52.2.31649
Xin Ji, Xinhua Wei, Anzhe Wang
A composite back-stepping sliding mode controller is explored in the paper to address the under-actuated, input saturated, and time-varying disturbances, as well as model-dependent issues that bother the path tracking control of unmanned agricultural tractors. Specifically, the path tracking error model is introduced. The extended state observers (ESO) with time-varying parameters are employed to handle the lump disturbances resulting from the external disturbances and model nonlinearity. A novel composite path tracking controller is proposed based on back-stepping and active disturbance rejection control and sliding mode control, whose effectiveness is elaborated by simulations and experiments. According to the results, the proposed controller, whose stability is elucidated in the appendix, outperforms the fuzzy pure pursuit control in reducing the lateral offset.
{"title":"A Novel Control Method for Unmanned Agricultural Tractors: Composite Back-stepping Sliding Mode Path Tracking","authors":"Xin Ji, Xinhua Wei, Anzhe Wang","doi":"10.5755/j01.itc.52.2.31649","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.31649","url":null,"abstract":"A composite back-stepping sliding mode controller is explored in the paper to address the under-actuated, input saturated, and time-varying disturbances, as well as model-dependent issues that bother the path tracking control of unmanned agricultural tractors. Specifically, the path tracking error model is introduced. The extended state observers (ESO) with time-varying parameters are employed to handle the lump disturbances resulting from the external disturbances and model nonlinearity. A novel composite path tracking controller is proposed based on back-stepping and active disturbance rejection control and sliding mode control, whose effectiveness is elaborated by simulations and experiments. According to the results, the proposed controller, whose stability is elucidated in the appendix, outperforms the fuzzy pure pursuit control in reducing the lateral offset.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"77 1","pages":"515-528"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73742254","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 : 2023-07-15DOI: 10.5755/j01.itc.52.2.33536
Branislav Radomirović, Vuk Jovanović, B. Nikolić, Sasa Stojanovic, K. Venkatachalam, M. Zivkovic, A. Njeguš, N. Bačanin, I. Strumberger
Due to the vast amounts of textual data available in various forms such as online content, social media comments, corporate data, public e-services and media data, text clustering has been experiencing rapid development. Text clustering involves categorizing and grouping similar content. It is a process of identifying significant patterns from unstructured textual data. Algorithms are being developed globally to extract useful and relevant information from large amounts of text data. Measuring the significance of content in documents to partition the collection of text data is one of the most important obstacles in text clustering. This study suggests utilizing an improved metaheuristics algorithm to fine-tune the K-means approach for text clustering task. The suggested technique is evaluated using the first 30 unconstrained test functions from the CEC2017 test-suite and six standard criterion text datasets. The simulation results and comparison with existing techniques demonstrate the robustness and supremacy of the suggested method.
{"title":"Text Document Clustering Approach by Improved Sine Cosine Algorithm","authors":"Branislav Radomirović, Vuk Jovanović, B. Nikolić, Sasa Stojanovic, K. Venkatachalam, M. Zivkovic, A. Njeguš, N. Bačanin, I. Strumberger","doi":"10.5755/j01.itc.52.2.33536","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.33536","url":null,"abstract":"Due to the vast amounts of textual data available in various forms such as online content, social media comments, corporate data, public e-services and media data, text clustering has been experiencing rapid development. Text clustering involves categorizing and grouping similar content. It is a process of identifying significant patterns from unstructured textual data. Algorithms are being developed globally to extract useful and relevant information from large amounts of text data. Measuring the significance of content in documents to partition the collection of text data is one of the most important obstacles in text clustering. This study suggests utilizing an improved metaheuristics algorithm to fine-tune the K-means approach for text clustering task. The suggested technique is evaluated using the first 30 unconstrained test functions from the CEC2017 test-suite and six standard criterion text datasets. The simulation results and comparison with existing techniques demonstrate the robustness and supremacy of the suggested method.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"57 1","pages":"541-561"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83122200","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 : 2023-07-15DOI: 10.5755/j01.itc.52.2.33208
M. Diwakaran, D. Surendran
Breast cancer is a major cause of death among women in both developed and underdeveloped countries. Early detection and diagnosis of breast cancer are crucial for patients to receive proper treatment and increase their chances of survival. To improve the automatic detection and diagnosis of breast cancer, a new deep learning model called “Breast Cancer Prognosis Based Transfer Learning (BCP-TL)” has been developed. This model uses transfer learning, which applies the knowledge gained from solving one problem to another relevant problem. The model is based on a pre-trained convolutional neural network (CNN) that extracts features from the mammographic image analysis society (MIAS) dataset. Four different CNN architectures were used in thismodel: AlexNet, Xception, ResNeXt, and Channel Boosted CNN. The performance of the model was evaluated using six metrics, including accuracy, sensitivity, specificity, precision, F1-score, and the area under the ROC curve (AUC). The combination of Xception and Channel Boosted CNN showed excellent performance. By combining essential features from multiple iterations, the Channel Boosted CNN can achieve higher accuracy in breast cancer diagnosis, with an overall accuracy of 98.96%. This highlights the potential of the BCP-TL model in effectively detecting and diagnosing breast cancer.
乳腺癌是发达国家和不发达国家妇女死亡的一个主要原因。乳腺癌的早期发现和诊断对于患者接受适当治疗和增加生存机会至关重要。为了提高乳腺癌的自动检测和诊断水平，提出了一种新的深度学习模型“基于乳腺癌预后的迁移学习(breast cancer Prognosis Based Transfer learning, BCP-TL)”。该模型使用迁移学习，将从解决一个问题中获得的知识应用于另一个相关问题。该模型基于预训练的卷积神经网络(CNN)，该网络从乳房x光图像分析学会(MIAS)数据集中提取特征。在这个模型中使用了四种不同的CNN架构:AlexNet、Xception、ResNeXt和Channel boosting CNN。采用精确性、敏感性、特异性、精密度、f1评分和ROC曲线下面积(AUC)等6个指标评价模型的性能。Xception和Channel boosting CNN的结合表现出了优异的性能。通过结合多次迭代的本质特征，Channel boosting CNN在乳腺癌诊断中可以达到更高的准确率，总体准确率达到98.96%。这突出了BCP-TL模型在有效检测和诊断乳腺癌方面的潜力。
{"title":"Breast Cancer Prognosis Based on Transfer Learning Techniques in Deep Neural Networks","authors":"M. Diwakaran, D. Surendran","doi":"10.5755/j01.itc.52.2.33208","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.33208","url":null,"abstract":"Breast cancer is a major cause of death among women in both developed and underdeveloped countries. Early detection and diagnosis of breast cancer are crucial for patients to receive proper treatment and increase their chances of survival. To improve the automatic detection and diagnosis of breast cancer, a new deep learning model called “Breast Cancer Prognosis Based Transfer Learning (BCP-TL)” has been developed. This model uses transfer learning, which applies the knowledge gained from solving one problem to another relevant problem. The model is based on a pre-trained convolutional neural network (CNN) that extracts features from the mammographic image analysis society (MIAS) dataset. Four different CNN architectures were used in thismodel: AlexNet, Xception, ResNeXt, and Channel Boosted CNN. The performance of the model was evaluated using six metrics, including accuracy, sensitivity, specificity, precision, F1-score, and the area under the ROC curve (AUC). The combination of Xception and Channel Boosted CNN showed excellent performance. By combining essential features from multiple iterations, the Channel Boosted CNN can achieve higher accuracy in breast cancer diagnosis, with an overall accuracy of 98.96%. This highlights the potential of the BCP-TL model in effectively detecting and diagnosing breast cancer.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"58 1","pages":"381-396"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77716257","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 : 2023-07-15DOI: 10.5755/j01.itc.52.2.33300
Yu Liu, Ningjie Zhou
Figure skating video jumping action is a complex combination action, which is difficult to recognize, and the recognition of jumping action can correct athletes’ technical errors, which is of great significance to improve athletes’ performance. Due to the recognition effect of figure skating video jumping action recognition algorithm is poor, we propose a figure skating video jumping action recognition algorithm using improved deep reinforcement learning in Internet of things (IoT). First, IoT technology is used to collect the figure skating video, the figure skating video target is detected, the human bone point features through the feature extraction network is obtained, and centralized processing is performed to complete the optimization of the extraction results. Second, the shallow STGCN network is improved to the DSTG dense connection network structure, based on which an improved deep reinforcement learning action recognition model is constructed, and the actionrecognition results are output through the deep network structure. Finally, a confidence fusion scheme is established to determine the final jumping action recognition result through the confidence is established. The results show that this paper effectively improves the accuracy of figure skating video jumping action recognition results, and the recognition quality is higher. It can be widely used in the field of figure skating action recognition, to improve the training effect of athletes.
{"title":"Jumping Action Recognition for Figure Skating Video in IoT Using Improved Deep Reinforcement Learning","authors":"Yu Liu, Ningjie Zhou","doi":"10.5755/j01.itc.52.2.33300","DOIUrl":"https://doi.org/10.5755/j01.itc.52.2.33300","url":null,"abstract":"Figure skating video jumping action is a complex combination action, which is difficult to recognize, and the recognition of jumping action can correct athletes’ technical errors, which is of great significance to improve athletes’ performance. Due to the recognition effect of figure skating video jumping action recognition algorithm is poor, we propose a figure skating video jumping action recognition algorithm using improved deep reinforcement learning in Internet of things (IoT). First, IoT technology is used to collect the figure skating video, the figure skating video target is detected, the human bone point features through the feature extraction network is obtained, and centralized processing is performed to complete the optimization of the extraction results. Second, the shallow STGCN network is improved to the DSTG dense connection network structure, based on which an improved deep reinforcement learning action recognition model is constructed, and the actionrecognition results are output through the deep network structure. Finally, a confidence fusion scheme is established to determine the final jumping action recognition result through the confidence is established. The results show that this paper effectively improves the accuracy of figure skating video jumping action recognition results, and the recognition quality is higher. It can be widely used in the field of figure skating action recognition, to improve the training effect of athletes.","PeriodicalId":54982,"journal":{"name":"Information Technology and Control","volume":"7 1","pages":"309-321"},"PeriodicalIF":1.1,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88428721","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：