Named Entity Recognition (NER) is a fundamental task that aids in the completion of other tasks such as text understanding, information retrieval and question answering in Natural Language Processing (NLP). In recent years, the use of a mix of character-word structure and dictionary information forChinese NER has been demonstrated to be effective. As a representative of hybrid models, Lattice-LSTM has obtained better benchmarking results in several publicly available Chinese NER datasets. However, Lattice-LSTM does not address the issue of long-distance entities or the detection of several entities with the same character. At the same time, the ambiguity of entity boundary information also leads to a decrease in the accuracy of embedding NER. This paper proposes ELCA: Enhanced Boundary Location for Chinese Named Entity Recognition Via Contextual Association, a method that solves the problem of long-distance dependent entities by using sentence-level position information. At the same time, it uses adaptive word convolution to overcome the problem of several entities sharing the same character. ELCA achieves the state-of-the-art outcomes in Chinese Word Segmentation and Chinese NER.
命名实体识别（NER）是一项基本任务，有助于完成自然语言处理（NLP）中的文本理解、信息检索和问题解答等其他任务。近年来，在中文 NER 中混合使用字词结构和词典信息已被证明是有效的。作为混合模型的代表，Lattice-LSTM 在几个公开的中文 NER 数据集中取得了较好的基准结果。然而，Lattice-LSTM 并没有解决远距离实体或多个同字实体的检测问题。同时，实体边界信息的模糊性也会导致嵌入 NER 的准确率下降。本文提出的 ELCA：通过上下文关联增强中文命名实体识别的边界定位，是一种利用句子级位置信息解决长距离依存实体问题的方法。同时，它还利用自适应词卷积克服了多个实体共享同一字符的问题。ELCA 在中文分词和中文近义词识别方面取得了最先进的成果。
{"title":"ELCA: Enhanced boundary location for Chinese named entity recognition via contextual association","authors":"Yizhao Wang, Shun Mao, Yuncheng Jiang","doi":"10.3233/ida-230383","DOIUrl":"https://doi.org/10.3233/ida-230383","url":null,"abstract":"Named Entity Recognition (NER) is a fundamental task that aids in the completion of other tasks such as text understanding, information retrieval and question answering in Natural Language Processing (NLP). In recent years, the use of a mix of character-word structure and dictionary information forChinese NER has been demonstrated to be effective. As a representative of hybrid models, Lattice-LSTM has obtained better benchmarking results in several publicly available Chinese NER datasets. However, Lattice-LSTM does not address the issue of long-distance entities or the detection of several entities with the same character. At the same time, the ambiguity of entity boundary information also leads to a decrease in the accuracy of embedding NER. This paper proposes ELCA: Enhanced Boundary Location for Chinese Named Entity Recognition Via Contextual Association, a method that solves the problem of long-distance dependent entities by using sentence-level position information. At the same time, it uses adaptive word convolution to overcome the problem of several entities sharing the same character. ELCA achieves the state-of-the-art outcomes in Chinese Word Segmentation and Chinese NER.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"23 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141745915","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}
Intrusion detection systems (IDSs) are essential elements of IT systems. Their key component is a classification module that continuously evaluates some features of the network traffic and identifies possible threats. Its efficiency is greatly affected by the right selection of the features to be monitored. Therefore, the identification of a minimal set of features that are necessary to safely distinguish malicious traffic from benign traffic is indispensable in the course of the development of an IDS. This paper presents the preprocessing and feature selection workflow as well as its results in the case of the CSE-CIC-IDS2018 on AWS dataset, focusing on five attack types. To identify the relevant features, six feature selection methods were applied, and the final ranking of the features was elaborated based on their average score. Next, several subsets of the features were formed based on different ranking threshold values, and each subset was tried with five classification algorithms to determine the optimal feature set for each attack type. During the evaluation, four widely used metrics were taken into consideration.
入侵检测系统（IDS）是 IT 系统的重要组成部分。其关键组件是一个分类模块，可持续评估网络流量的某些特征并识别可能的威胁。正确选择要监控的特征对其效率影响很大。因此，在开发 IDS 的过程中，必须确定一组最基本的特征，以便安全地区分恶意流量和良性流量。本文以 AWS 上的 CSE-CIC-IDS2018 数据集为例，介绍了预处理和特征选择工作流程及其结果，重点关注五种攻击类型。为了识别相关特征，采用了六种特征选择方法，并根据平均得分对特征进行了最终排序。接下来，根据不同的排名阈值形成了多个特征子集，并用五种分类算法对每个子集进行了尝试，以确定每种攻击类型的最佳特征集。在评估过程中，考虑了四个广泛使用的指标。
{"title":"Identifying relevant features of CSE-CIC-IDS2018 dataset for the development of an intrusion detection system","authors":"László Göcs, Zsolt Csaba Johanyák","doi":"10.3233/ida-230264","DOIUrl":"https://doi.org/10.3233/ida-230264","url":null,"abstract":"Intrusion detection systems (IDSs) are essential elements of IT systems. Their key component is a classification module that continuously evaluates some features of the network traffic and identifies possible threats. Its efficiency is greatly affected by the right selection of the features to be monitored. Therefore, the identification of a minimal set of features that are necessary to safely distinguish malicious traffic from benign traffic is indispensable in the course of the development of an IDS. This paper presents the preprocessing and feature selection workflow as well as its results in the case of the CSE-CIC-IDS2018 on AWS dataset, focusing on five attack types. To identify the relevant features, six feature selection methods were applied, and the final ranking of the features was elaborated based on their average score. Next, several subsets of the features were formed based on different ranking threshold values, and each subset was tried with five classification algorithms to determine the optimal feature set for each attack type. During the evaluation, four widely used metrics were taken into consideration.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"2017 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2024-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141151739","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}
Da Tong, Shudong Chen, Rong Ma, Donglin Qi, Yong Yu
Knowledge graph embedding (KGE) is typically used for link prediction to automatically predict missing links in knowledge graphs. Current KGE models are mainly based on complicated mathematical associations, which are highly expressive but ignore the uniformity behind the classical bilinear translational model TransE, a model that embeds all entities of knowledge graphs in a uniform space, enabling accurate embeddings. This study analyses the uniformity of TransE and proposes a novel KGE model called ConvUs that follows uniformity with expressiveness. Based on the convolution neural network (CNN), ConvUs proposes constraints on convolution filter values and employs a multi-layer, multi-scale CNN architecture with a non-parametric L2 norm-based scoring function for the calculation of triple scores. This addresses potential uniformity-related issues in existing CNN-based KGE models, allowing ConvUs to maintain a uniform embedding space while benefiting from the powerful expressiveness of CNNs. Furthermore, circular convolution is applied to alleviate the potential orderliness contradictions, making ConvUs more suitable for conducting uniform space KGE. Our model outperformed the base model ConvKB and several baselines on the link prediction benchmark WN18RR and FB15k-237, demonstrating strong applicability and generalization and indicating that the uniformity of embedding space with high expressiveness enables more efficient knowledge graph embeddings.
{"title":"Knowledge graph embedding in a uniform space","authors":"Da Tong, Shudong Chen, Rong Ma, Donglin Qi, Yong Yu","doi":"10.3233/ida-227123","DOIUrl":"https://doi.org/10.3233/ida-227123","url":null,"abstract":"Knowledge graph embedding (KGE) is typically used for link prediction to automatically predict missing links in knowledge graphs. Current KGE models are mainly based on complicated mathematical associations, which are highly expressive but ignore the uniformity behind the classical bilinear translational model TransE, a model that embeds all entities of knowledge graphs in a uniform space, enabling accurate embeddings. This study analyses the uniformity of TransE and proposes a novel KGE model called ConvUs that follows uniformity with expressiveness. Based on the convolution neural network (CNN), ConvUs proposes constraints on convolution filter values and employs a multi-layer, multi-scale CNN architecture with a non-parametric L2 norm-based scoring function for the calculation of triple scores. This addresses potential uniformity-related issues in existing CNN-based KGE models, allowing ConvUs to maintain a uniform embedding space while benefiting from the powerful expressiveness of CNNs. Furthermore, circular convolution is applied to alleviate the potential orderliness contradictions, making ConvUs more suitable for conducting uniform space KGE. Our model outperformed the base model ConvKB and several baselines on the link prediction benchmark WN18RR and FB15k-237, demonstrating strong applicability and generalization and indicating that the uniformity of embedding space with high expressiveness enables more efficient knowledge graph embeddings.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"5 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2024-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139927130","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}
Consultative leadership is a democratic style that deliberately incorporates employees into organizational management and decision-making to increase employees’ feelings of ownership and align their objectives with company objectives. As a result, during their everyday tasks, leaders constantly utilize “consultation management” for their staff. As examples, consider how to coordinate reports, communicate key ideas, and use a variety of flexible promotion strategies. This study investigates the role of consultive leadership on administrative development in developing countries. For this reason, this study has applied a questionnaire to take the respondents’ opinions in the Iraqi ministry of interior affairs. Using the Likert scale has provided quantitative value for the qualitative study. For this reason, questionnaires were provided, and this study’s results showed a positive correlation between consultative leadership and administrative development. As a result, the organization’s leader has more chances to administer the organization successfully than a manager or an unofficial leader who lacks status power.
{"title":"The role of consultative leadership on administrative development","authors":"Ayas Mohammed Rasheed Omar, Khairi Ali Auso","doi":"10.3233/ida-237448","DOIUrl":"https://doi.org/10.3233/ida-237448","url":null,"abstract":"Consultative leadership is a democratic style that deliberately incorporates employees into organizational management and decision-making to increase employees’ feelings of ownership and align their objectives with company objectives. As a result, during their everyday tasks, leaders constantly utilize “consultation management” for their staff. As examples, consider how to coordinate reports, communicate key ideas, and use a variety of flexible promotion strategies. This study investigates the role of consultive leadership on administrative development in developing countries. For this reason, this study has applied a questionnaire to take the respondents’ opinions in the Iraqi ministry of interior affairs. Using the Likert scale has provided quantitative value for the qualitative study. For this reason, questionnaires were provided, and this study’s results showed a positive correlation between consultative leadership and administrative development. As a result, the organization’s leader has more chances to administer the organization successfully than a manager or an unofficial leader who lacks status power.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"1 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139201563","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}
Anum Yasmin, Wasi Haider, Ali Daud, Ameena T Banjar
Crowd-Sourced software development (CSSD) is getting a good deal of attention from the software and research community in recent times. One of the key challenges faced by CSSD platforms is the task selection mechanism which in practice, contains no intelligent scheme. Rather, rule-of-thumb or intuition strategies are employed, leading to biasness and subjectivity. Effort considerations on crowdsourced tasks can offer good foundation for task selection criteria but are not much investigated. Software development effort estimation (SDEE) is quite prevalent domain in software engineering but only investigated for in-house development. For open-sourced or crowdsourced platforms, it is rarely explored. Moreover, Machine learning (ML) techniques are overpowering SDEE with a claim to provide more accurate estimation results. This work aims to conjoin ML-based SDEE to analyze development effort measures on CSSD platform. The purpose is to discover development-oriented features for crowdsourced tasks and analyze performance of ML techniques to find best estimation model on CSSD dataset. TopCoder is selected as target CSSD platform for the study. TopCoder’s development tasks data with development-centric features are extracted, leading to statistical, regression and correlation analysis to justify features’ significance. For effort estimation, 10 ML families with 2 respective techniques are applied to get broader aspect of estimation. Five performance metrices (MSE, RMSE, MMRE, MdMRE, Pred (25) and Welch’s statistical test are incorporated to judge the worth of effort estimation model’s performance. Data analysis results show that selected features of TopCoder pertain reasonable model significance, regression, and correlation measures. Findings of ML effort estimation depicted that best results for TopCoder dataset can be acquired by linear, non-linear regression and SVM family models. To conclude, the study identified the most relevant development features for CSSD platform, confirmed by in-depth data analysis. This reflects careful selection of effort estimation features to offer good basis of accurate ML estimate.
众包软件开发（CSSD）近来受到软件和研究界的广泛关注。CSSD 平台面临的主要挑战之一是任务选择机制。相反，由于采用了经验法则或直觉策略，导致了偏差和主观性。众包任务的努力考虑因素可以为任务选择标准提供良好的基础，但这方面的研究并不多。软件开发工作量估算（SDEE）是软件工程中相当普遍的领域，但只针对内部开发进行过研究。对于开源或众包平台，很少有人进行过研究。此外，机器学习（ML）技术正在取代 SDEE，声称能提供更准确的估算结果。这项工作旨在结合基于 ML 的 SDEE，分析 CSSD 平台上的开发工作量。其目的是发现众包任务的开发导向特征，并分析 ML 技术的性能，从而找到 CSSD 数据集上的最佳估算模型。本研究选择 TopCoder 作为目标 CSSD 平台。研究人员从 TopCoder 的开发任务数据中提取了以开发为中心的特征，并进行了统计、回归和相关分析，以证明特征的重要性。在工作量估算方面，应用了 10 个 ML 族和 2 种不同的技术，以获得更广泛的估算结果。五种性能指标（MSE、RMSE、MMRE、MdMRE、Pred (25) 和韦尔奇统计检验）用于判断努力估算模型的性能价值。数据分析结果表明，TopCoder 的选定功能与合理的模型显著性、回归和相关测量有关。ML 努力估算的结果表明，TopCoder 数据集的最佳结果可通过线性、非线性回归和 SVM 系列模型获得。总之，通过深入的数据分析，本研究确定了与 CSSD 平台最相关的开发特征。这反映了对努力估算特征的精心选择，为准确的 ML 估算提供了良好的基础。
{"title":"Machine learning based software effort estimation using development-centric features for crowdsourcing platform","authors":"Anum Yasmin, Wasi Haider, Ali Daud, Ameena T Banjar","doi":"10.3233/ida-237366","DOIUrl":"https://doi.org/10.3233/ida-237366","url":null,"abstract":"Crowd-Sourced software development (CSSD) is getting a good deal of attention from the software and research community in recent times. One of the key challenges faced by CSSD platforms is the task selection mechanism which in practice, contains no intelligent scheme. Rather, rule-of-thumb or intuition strategies are employed, leading to biasness and subjectivity. Effort considerations on crowdsourced tasks can offer good foundation for task selection criteria but are not much investigated. Software development effort estimation (SDEE) is quite prevalent domain in software engineering but only investigated for in-house development. For open-sourced or crowdsourced platforms, it is rarely explored. Moreover, Machine learning (ML) techniques are overpowering SDEE with a claim to provide more accurate estimation results. This work aims to conjoin ML-based SDEE to analyze development effort measures on CSSD platform. The purpose is to discover development-oriented features for crowdsourced tasks and analyze performance of ML techniques to find best estimation model on CSSD dataset. TopCoder is selected as target CSSD platform for the study. TopCoder’s development tasks data with development-centric features are extracted, leading to statistical, regression and correlation analysis to justify features’ significance. For effort estimation, 10 ML families with 2 respective techniques are applied to get broader aspect of estimation. Five performance metrices (MSE, RMSE, MMRE, MdMRE, Pred (25) and Welch’s statistical test are incorporated to judge the worth of effort estimation model’s performance. Data analysis results show that selected features of TopCoder pertain reasonable model significance, regression, and correlation measures. Findings of ML effort estimation depicted that best results for TopCoder dataset can be acquired by linear, non-linear regression and SVM family models. To conclude, the study identified the most relevant development features for CSSD platform, confirmed by in-depth data analysis. This reflects careful selection of effort estimation features to offer good basis of accurate ML estimate.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"29 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139201259","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}
Wilawan Yathongkhum, Y. Laosiritaworn, Jakramate Bootkrajang, Pucktada Treeratpituk, Jeerayut Chaijaruwanich
A large amount of economic and financial news is now accessible through various news websites and social media platforms. Categorizing them into appropriate categories can be advantageous for various tasks, such as sentiment analysis and news-based market prediction. Unfortunately, news headlines categories may contain ambiguities due to the subjective nature of label assignment by authors or publishers. Consequently, achieving precise classification of news can be time-consuming and still reliant on human expertise. To tackle this challenging task, we proposed a hybrid approach to enhance the performance of economic and financial news classification. This approach combines baseline classifiers with a novel method called the Category Associated Feature Set (CAFS) classifier. CAFS transforms text input from the lexicon-space into the entity-space and discovers associations between entities and classes, akin to association rule learning. Experimental results on three datasets demonstrated that the proposed method is comparable to existing approaches and exhibits a significant improvement in the classification results for out-of-domain datasets. Additionally, employing CAFS in tandem with the existing text classification baselines can provide a general categorizer for distinguishing news categories across various sources without the need for extensive fine-tuning of the parameters associated with those classification baselines. This confirms that utilizing CAFS in a hybrid approach is appropriate and suitable for economic and financial news classification.
{"title":"Economic and financial news hybrid- classification based on category-associated feature set","authors":"Wilawan Yathongkhum, Y. Laosiritaworn, Jakramate Bootkrajang, Pucktada Treeratpituk, Jeerayut Chaijaruwanich","doi":"10.3233/ida-237373","DOIUrl":"https://doi.org/10.3233/ida-237373","url":null,"abstract":"A large amount of economic and financial news is now accessible through various news websites and social media platforms. Categorizing them into appropriate categories can be advantageous for various tasks, such as sentiment analysis and news-based market prediction. Unfortunately, news headlines categories may contain ambiguities due to the subjective nature of label assignment by authors or publishers. Consequently, achieving precise classification of news can be time-consuming and still reliant on human expertise. To tackle this challenging task, we proposed a hybrid approach to enhance the performance of economic and financial news classification. This approach combines baseline classifiers with a novel method called the Category Associated Feature Set (CAFS) classifier. CAFS transforms text input from the lexicon-space into the entity-space and discovers associations between entities and classes, akin to association rule learning. Experimental results on three datasets demonstrated that the proposed method is comparable to existing approaches and exhibits a significant improvement in the classification results for out-of-domain datasets. Additionally, employing CAFS in tandem with the existing text classification baselines can provide a general categorizer for distinguishing news categories across various sources without the need for extensive fine-tuning of the parameters associated with those classification baselines. This confirms that utilizing CAFS in a hybrid approach is appropriate and suitable for economic and financial news classification.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"5 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139202492","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}
Nowadays, the massive industrial data has effectively improved the performance of the data-driven deep learning Remaining Useful Life (RUL) prediction method. However, there are still problems of assigning fixed weights to features and only coarse-grained consideration at the sequence level. This paper proposes a Transformer-based end-to-end feature-level mask self-supervised learning method for RUL prediction. First, by proposing a fine-grained feature-level mask self-supervised learning method, the data at different time points under all features in a time window is sent to two parallel learning streams with and without random masks. The model can learn more fine-grained degradation information by comparing the information extracted by the two parallel streams. Instead of assigning fixed weights to different features, the abstract information extracted through the above process is invariable correlations between features, which has a good generalization to various situations under different working conditions. Then, the extracted information is encoded and decoded again using an asymmetric structure, and a fully connected network is used to build a mapping between the extracted information and the RUL. We conduct experiments on the public C-MAPSS datasets and show that the proposed method outperforms the other methods, and its advantages are more obvious in complex multi-working conditions.
{"title":"A feature-level mask self-supervised assisted learning approach based on transformer for remaining useful life prediction","authors":"Bing Xue, Xin Gao, Shuwei Zhang, Ning Wang, Shiyuan Fu, Jiahao Yu, Guangyao Zhang, Zijian Huang","doi":"10.3233/ida-227099","DOIUrl":"https://doi.org/10.3233/ida-227099","url":null,"abstract":"Nowadays, the massive industrial data has effectively improved the performance of the data-driven deep learning Remaining Useful Life (RUL) prediction method. However, there are still problems of assigning fixed weights to features and only coarse-grained consideration at the sequence level. This paper proposes a Transformer-based end-to-end feature-level mask self-supervised learning method for RUL prediction. First, by proposing a fine-grained feature-level mask self-supervised learning method, the data at different time points under all features in a time window is sent to two parallel learning streams with and without random masks. The model can learn more fine-grained degradation information by comparing the information extracted by the two parallel streams. Instead of assigning fixed weights to different features, the abstract information extracted through the above process is invariable correlations between features, which has a good generalization to various situations under different working conditions. Then, the extracted information is encoded and decoded again using an asymmetric structure, and a fully connected network is used to build a mapping between the extracted information and the RUL. We conduct experiments on the public C-MAPSS datasets and show that the proposed method outperforms the other methods, and its advantages are more obvious in complex multi-working conditions.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"17 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139200197","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}
The performance of Adaboost is highly sensitive to noisy and outlier samples. This is therefore the weights of these samples are exponentially increased in successive rounds. In this paper, three novel schemes are proposed to hunt the corrupted samples and eliminate them through the training process. The methods are: I) a hybrid method based on K-means clustering and K-nearest neighbor, II) a two-layer Adaboost, and III) soft margin support vector machines. All of these solutions are compared to the standard Adaboost on thirteen Gunnar Raetsch’s datasets under three levels of class-label noise. To test the proposed method on a real application, electroencephalography (EEG) signals of 20 schizophrenic patients and 20 age-matched control subjects, are recorded via 20 channels in the idle state. Several features including autoregressive coefficients, band power and fractal dimension are extracted from EEG signals of all participants. Sequential feature subset selection technique is adopted to select the discriminative EEG features. Experimental results imply that exploiting the proposed hunting techniques enhance the Adaboost performance as well as alleviating its robustness against unconfident and noisy samples over Raetsch benchmark and EEG features of the two groups.
Adaboost 的性能对噪声样本和离群样本非常敏感。因此，这些样本的权重会在连续几轮中呈指数增长。本文提出了三种新方案，通过训练过程猎取并消除损坏的样本。这些方法是I) 基于 K 均值聚类和 K 近邻的混合方法；II) 两层 Adaboost；III) 软边际支持向量机。所有这些解决方案都在 Gunnar Raetsch 的 13 个数据集上与标准 Adaboost 方法进行了比较，这些数据集具有三种等级的类标签噪声。为了在实际应用中测试所提出的方法，在空闲状态下通过 20 个通道记录了 20 名精神分裂症患者和 20 名年龄匹配的对照受试者的脑电图（EEG）信号。从所有参与者的脑电信号中提取了一些特征，包括自回归系数、频带功率和分形维度。采用序列特征子集选择技术来选择具有区分性的脑电图特征。实验结果表明，与 Raetsch 基准和两组脑电图特征相比，利用所提出的狩猎技术提高了 Adaboost 的性能，并减轻了其对无把握样本和噪声样本的鲁棒性。
{"title":"Enhancing Adaboost performance in the presence of class-label noise: A comparative study on EEG-based classification of schizophrenic patients and benchmark datasets","authors":"O. R. Pouya, Reza Boostani, M. Sabeti","doi":"10.3233/ida-227125","DOIUrl":"https://doi.org/10.3233/ida-227125","url":null,"abstract":"The performance of Adaboost is highly sensitive to noisy and outlier samples. This is therefore the weights of these samples are exponentially increased in successive rounds. In this paper, three novel schemes are proposed to hunt the corrupted samples and eliminate them through the training process. The methods are: I) a hybrid method based on K-means clustering and K-nearest neighbor, II) a two-layer Adaboost, and III) soft margin support vector machines. All of these solutions are compared to the standard Adaboost on thirteen Gunnar Raetsch’s datasets under three levels of class-label noise. To test the proposed method on a real application, electroencephalography (EEG) signals of 20 schizophrenic patients and 20 age-matched control subjects, are recorded via 20 channels in the idle state. Several features including autoregressive coefficients, band power and fractal dimension are extracted from EEG signals of all participants. Sequential feature subset selection technique is adopted to select the discriminative EEG features. Experimental results imply that exploiting the proposed hunting techniques enhance the Adaboost performance as well as alleviating its robustness against unconfident and noisy samples over Raetsch benchmark and EEG features of the two groups.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"32 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139199232","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}
Sung Hwan Jeon, Hyeonguk Lee, Jihye Park, Sungzoon Cho
Attempts to express information from various documents in graph form are rapidly increasing. The speed and volume in which these documents are being generated call for an automated process, based on machine learning techniques, for cost-effective and timely analysis. Past studies responded to such needs by building knowledge graphs or technology trees from the bibliographic information of documents, or by relying on text mining techniques in order to extract keywords and/or phrases. While these approaches provide an intuitive glance into the technological hotspots or the key features of the select field, there still is room for improvement, especially in terms of recognizing the same entities appearing in different forms so as to interconnect closely related technological concepts properly. In this paper, we propose to build a patent knowledge network using the United States Patent and Trademark Office (USPTO) patent filings for the semiconductor device sector by fine-tuning Huggingface’s named entity recognition (NER) model with our novel edge weight updating neural network. For the named entity normalization, we employ edge weight updating neural network with positive and negative candidates that are chosen by substring matching techniques. Experiment results show that our proposed approach performs very competitively against the conventional keyword extraction models frequently employed in patent analysis, especially for the named entity normalization (NEN) and document retrieval tasks. By grouping entities with named entity normalization model, the resulting knowledge graph achieves higher scores in retrieval tasks. We also show that our model is robust to the out-of-vocabulary problem by employing the fine-tuned BERT NER model.
以图表形式表达各种文件信息的尝试正在迅速增加。这些文档生成的速度和数量要求基于机器学习技术的自动化流程，以便进行经济高效的及时分析。过去的研究通过从文件的书目信息中建立知识图谱或技术树，或依靠文本挖掘技术提取关键词和/或短语来满足这些需求。虽然这些方法能让人直观地了解技术热点或所选领域的关键特征，但仍有改进的余地，特别是在识别以不同形式出现的相同实体，从而将密切相关的技术概念正确地联系起来方面。在本文中，我们建议利用美国专利商标局（USPTO）半导体设备领域的专利申请，通过微调 Huggingface 的命名实体识别（NER）模型和我们新颖的边缘权重更新神经网络来构建专利知识网络。在命名实体规范化方面，我们采用边缘权重更新神经网络，通过子串匹配技术选择正负候选实体。实验结果表明，与专利分析中常用的传统关键词提取模型相比，我们提出的方法具有很强的竞争力，尤其是在命名实体规范化（NEN）和文档检索任务方面。通过命名实体归一化模型对实体进行分组，生成的知识图谱在检索任务中获得了更高的分数。我们还表明，通过采用微调 BERT NER 模型，我们的模型对词汇表外问题具有很强的鲁棒性。
{"title":"Building knowledge graphs from technical documents using named entity recognition and edge weight updating neural network with triplet loss for entity normalization","authors":"Sung Hwan Jeon, Hyeonguk Lee, Jihye Park, Sungzoon Cho","doi":"10.3233/ida-227129","DOIUrl":"https://doi.org/10.3233/ida-227129","url":null,"abstract":"Attempts to express information from various documents in graph form are rapidly increasing. The speed and volume in which these documents are being generated call for an automated process, based on machine learning techniques, for cost-effective and timely analysis. Past studies responded to such needs by building knowledge graphs or technology trees from the bibliographic information of documents, or by relying on text mining techniques in order to extract keywords and/or phrases. While these approaches provide an intuitive glance into the technological hotspots or the key features of the select field, there still is room for improvement, especially in terms of recognizing the same entities appearing in different forms so as to interconnect closely related technological concepts properly. In this paper, we propose to build a patent knowledge network using the United States Patent and Trademark Office (USPTO) patent filings for the semiconductor device sector by fine-tuning Huggingface’s named entity recognition (NER) model with our novel edge weight updating neural network. For the named entity normalization, we employ edge weight updating neural network with positive and negative candidates that are chosen by substring matching techniques. Experiment results show that our proposed approach performs very competitively against the conventional keyword extraction models frequently employed in patent analysis, especially for the named entity normalization (NEN) and document retrieval tasks. By grouping entities with named entity normalization model, the resulting knowledge graph achieves higher scores in retrieval tasks. We also show that our model is robust to the out-of-vocabulary problem by employing the fine-tuned BERT NER model.","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"173 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139208091","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}
Cairong Yan, Xiaoke Li, Ran Tao, Zhaohui Zhang, Yongquan Wan
Extracting more information from feature interactions is essential to improve click-through rate (CTR) prediction accuracy. Although deep learning technology can help capture high-order feature interactions, the combination of features lacks interpretability. In this paper, we propose a multi-semantic feature interaction learning network (MeFiNet), which utilizes convolution operations to map feature interactions to multi-semantic spaces to improve their expressive ability and uses an improved Squeeze & Excitation method based on SENet to learn the importance of these interactions in different semantic spaces. The Squeeze operation helps to obtain the global importance distribution of semantic spaces, and the Excitation operation helps to dynamically re-assign the weights of semantic features so that both semantic diversity and feature diversity are considered in the model. The generated multi-semantic feature interactions are concatenated with the original feature embeddings and input into a deep learning network. Experiments on three public datasets demonstrate the effectiveness of the proposed model. Compared with state-of-the-art methods, the model achieves excellent performance (+0.18% in AUC and -0.34% in LogLoss VS DeepFM; +0.19% in AUC and -0.33% in LogLoss VS FiBiNet).
{"title":"MeFiNet: Modeling multi-semantic convolution-based feature interactions for CTR prediction","authors":"Cairong Yan, Xiaoke Li, Ran Tao, Zhaohui Zhang, Yongquan Wan","doi":"10.3233/ida-227113","DOIUrl":"https://doi.org/10.3233/ida-227113","url":null,"abstract":"Extracting more information from feature interactions is essential to improve click-through rate (CTR) prediction accuracy. Although deep learning technology can help capture high-order feature interactions, the combination of features lacks interpretability. In this paper, we propose a multi-semantic feature interaction learning network (MeFiNet), which utilizes convolution operations to map feature interactions to multi-semantic spaces to improve their expressive ability and uses an improved Squeeze & Excitation method based on SENet to learn the importance of these interactions in different semantic spaces. The Squeeze operation helps to obtain the global importance distribution of semantic spaces, and the Excitation operation helps to dynamically re-assign the weights of semantic features so that both semantic diversity and feature diversity are considered in the model. The generated multi-semantic feature interactions are concatenated with the original feature embeddings and input into a deep learning network. Experiments on three public datasets demonstrate the effectiveness of the proposed model. Compared with state-of-the-art methods, the model achieves excellent performance (+0.18% in AUC and -0.34% in LogLoss VS DeepFM; +0.19% in AUC and -0.33% in LogLoss VS FiBiNet).","PeriodicalId":50355,"journal":{"name":"Intelligent Data Analysis","volume":"14 6","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139197290","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：