[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-273346-105":59,"doc-detail-273346-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","cpc-cms-cognitive-pairwise-comparison-classification-model-selection-framework-for-document-level-sentiment-analysis","CPC-CMS - Cognitive Pairwise Comparison Classification Model Selection Framework for Document-level Sentiment Analysis","","This study proposes the Cognitive Pairwise Comparison Classification Model Selection (CPC-CMS) framework for document-level sentiment analysis. The CPC, driven by expert judgment, assigns weights to multiple evaluation criteria, including accuracy, precision, recall, F1-score, specificity, MCC, Cohen's Kappa, and efficiency. Baseline classifiers span NB, LSVC, Random Forest, Logistic Regression, XGBoost, LSTM, and ALBERT. A weighted decision matrix selects the best model using criteria scores and CPC-derived weights. Experiments on three open social media datasets show that excluding time, ALBERT is best across datasets, while including time yields no consistent winner. Comparisons with AHP, TOPSIS, and MOORA vary in aggregation values and ranks. Spearman sensitivity testing supports robustness, and CPC-CMS generalizes to other classification domains.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/cpc-cms-cognitive-pairwise-comparison-classification-model-selection-framework-for-document-level-sentiment-analysis/273346/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/cpc-cms-cognitive-pairwise-comparison-classification-model-selection-framework-for-document-level-sentiment-analysis/273346.png","ImageObject",300,407,{"name":92,"@type":93},"Jiven","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-29","2026-09-15",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",11,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What does the CPC-CMS framework use to select a document-level sentiment model?","Question",{"text":113,"@type":114},"It uses Cognitive Pairwise Comparison (CPC) to determine weights for multiple evaluation criteria, then applies a weighted decision matrix to choose the best classification model based on criteria scores.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"Which classification models are used as baseline candidates in CPC-CMS?",{"text":118,"@type":114},"The framework evaluates Naive Bayes, Linear Support Vector Classification, Random Forest, Logistic Regression, Extreme Gradient Boosting (XGBoost), Long Short-Term Memory (LSTM), and ALBERT.",{"name":120,"@type":111,"acceptedAnswer":121},"What do the dataset experiments indicate about ALBERT and the effect of time?",{"text":122,"@type":114},"Excluding the time factor, ALBERT performs best across all three datasets. When the time factor is included, no single model consistently outperforms the others.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},273346,1789470984,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":36,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":130,"read_time":144,"preview_page_count":4},1099513958607,"https://ap-avatar.wpscdn.com/avatar/100002390cf8733938c?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778829742770036399","CPC-CMS: Cognitive Pairwise Comparison Classification Model Selection Framework for Document-level Sentiment  \nAnalysis  \nJianfei LI1, Kevin Kam Fung YUEN2*,1  \n1Department of Computing, Faculty of Computer and Mathematical Sciences The Hong Kong Polytechnic University, Hong Kong SAR, China  \n2*School of Science, Monash University Malaysia,  \nJalan Lagoon Selatan, Malaysia  \n*[kevinkf.yuen@gmail.com](kevinkf.yuen@gmail.com), [kevin.yuen@monash.edu](kevin.yuen@monash.edu)  \n*[https://orcid.org/0000-0003-1497-2575](https://orcid.org/0000-0003-1497-2575)  \nAbstract  \nThis study proposes the Cognitive Pairwise Comparison Classification Model Selection (CPC-CMS) framework for document-level sentiment analysis. The CPC, based on expert knowledge judgment, is used to calculate the weights of evaluation criteria, including accuracy, precision, recall, F1-score, specificity, Matthews Correlation Coefficient (MCC), Cohen's Kappa (Kappa), and efficiency. Naive Bayes (NB), Linear Support Vector Classification (LSVC), Random Forest, Logistic Regression, Extreme Gradient Boosting (XGBoost), Long Short-Term Memory (LSTM), and A Lite Bidirectional Encoder Representations from Transformers (ALBERT) are chosen as classification baseline models. A weighted decision matrix consisting of classification evaluation scores with respect to criteria weights is formed to select the best classification model for a classification problem. Three open social media datasets are used to demonstrate the feasibility of the proposed CPC-CMS. Based on our simulation, for evaluation results excluding the time factor, ALBERT performs best across all three datasets; if the time factor is included, no single model consistently outperforms the others. Through comparison, these conclusions are also supported by other aggregation and ranking methods, including Analytic Hierarchy Process (AHP), Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and Multi-Objective Optimization by Ratio Analysis (MOORA), although aggregation values and ranks may vary. A sensitivity analysis using Spearman's Rank Correlation Test demonstrates the robustness of the proposed CPC-CMS framework. The CPC-CMS can be applied to other classification applications in various domains.  \nKeywords: Sentiment Analysis, Multi-Criteria Decision Making, NLP Model Selection, Text Classification, Cognitive Pairwise Comparison.  \n1. Introduction  \nSentiment analysis is a subfield of natural language processing, aiming to extract and analyze sentiments and viewpoints from textual information to facilitate computational research on the attitudes and sentiments  \nof entities such as topics, individuals, issues, and products [1] . The Internet and mobile phones have occupied a large part of people's lives, leading to an explosive growth in the number of platforms such as Twitter and online comments [2] . Automatically analyzing large volumes of comments quickly, efficiently, and at low cost to make corresponding decisions has generated great interest in various fields (e.g., e-commerce, healthcare, government departments, etc.) [3] . For example, in the field of healthcare, Baker et al. [4] reliably analyzed people's responses to colon cancer using the GRU model and made predictions for the disease. In the business sector, Rognone et al. [5] investigated the impact of news sentiment on the returns, trading volumes, and volatility of Bitcoin and traditional currencies.  \nMany sentiment analysis problems can be treated as classification problems, and they can be performed at three levels: the document level, the sentence level, and the aspect level [6] . For document level, a document containing all sentences is regarded as an independent object, expressing an overall opinion. For sentence level, a subjectivity label, either subjective or objective, is given to each sentence. Since objective sentences often do not convey any viewpoints, sentiment classification determines the sentiments (e.g","cbCaimltGRJyjDCT","https://ap.wps.com/l/cbCaimltGRJyjDCT","pdf",1696767,"English","# 1. Introduction\n## Sentiment analysis as classification at multiple levels\n## Model evaluation, selection, and the need for multi-criteria aggregation\n## CPC-CMS framework overview and cognitive comparison motivation","[{\"question\":\"What does the CPC-CMS framework use to select a document-level sentiment model?\",\"answer\":\"It uses Cognitive Pairwise Comparison (CPC) to determine weights for multiple evaluation criteria, then applies a weighted decision matrix to choose the best classification model based on criteria scores.\"},{\"question\":\"Which classification models are used as baseline candidates in CPC-CMS?\",\"answer\":\"The framework evaluates Naive Bayes, Linear Support Vector Classification, Random Forest, Logistic Regression, Extreme Gradient Boosting (XGBoost), Long Short-Term Memory (LSTM), and ALBERT.\"},{\"question\":\"What do the dataset experiments indicate about ALBERT and the effect of time?\",\"answer\":\"Excluding the time factor, ALBERT performs best across all three datasets. When the time factor is included, no single model consistently outperforms the others.\"}]","CPC-CMS - Cognitive Pairwise Comparison Classification Model Selection Framework for Document-level Sentiment Analysis | PDF",101]