This paper aims to conduct an in -depth study on the sentiment analysis of emails based on Graph Convolutional Networks (GCNs) (our model) and proposes a joint modeling method that integrates sentiment and behavioral features. The research presented in this paper makes use of a variety of open -source datasets, including, but not limited to, files from Enron, SpamAssassin, and Ama zon Reviews, to mention just a few. In order to do preprocessing on the dataset, noise should be removed, the dataset should be broken up into tokens, and stop words should be eliminated. This will guarantee that the data is of the highest possible quality at all times. This is followed by the use of a multitude of baseline models, including Bag of Words (BoW) , Term Frequency -Inverse Document Frequency ( TF -IDF ), Long Short -Term Memory ( LSTM ), Bidirectional Encoder Representations from Transformers ( BERT ), and others, in order to validate the job of the sentiment analysis. When it came to the exercise of categorizing sentiments, we performed far better than the other baseline models. In terms of accuracy, precis ion, recall, and F1 value, our Area Under the Cu rve (AUC) was 0.94, our accuracy rate was 92.5%, our precision rate was 90.3%, and our recall rate was 86.7%. In comparison to prior models, we improved the F1 value by 2.3% and made the integrated modeling of emotion and behavior 3.7% more accurate. In the last sec tion of the research project, the researchers investigate potential applications of the GCN -based sentiment analysis model in the real world, with the goal of enhancing people ’s motivation and happiness. According to the findings of the research, our approach is capable of accurately identifying changes in state of mind and conduct in the content of emails. In real scenarios, user satisfaction increased by 15% and engagement increased by 20%. Therefore, our model has broad application prospects, e specially in personalized recommendation systems and intelligent service optimization .

[1] Al -Qurishi M., Alhuzami S., AlRubaian M., Hossain M., and et al., “User Profiling for Big Social Media Data Using Standing Ovation Model,” Multimedia Tools and Applications , vol. 77, no. 9, pp. 11179 -11201, 2018. https://doi.org/10.1007/s11042 -017 -5402 -6

[2] Cesconetto J., Silva L., Bortoluzzi F., Navarro - Cáceres M., and et al., “PRIPRO -Privacy Profiles: User Profiling Management for Smart Environments,” Electronics , vol. 9, no. 9, pp. 1 - 22, 2020. https://doi.org/10.3390/electronics9091519

[3] Del Valle -Cano G., Quijano -Sánchez L., Liberatore F., and Gómez J., “SocialHaterBERT: A Dichotomous Approach for Automatically Detecting Hate Speech on Twitter Through Textual Analysis and User Profiles,” Expert Systems with Applications , vol. 216, pp. 1 -17, 2023. https://doi.org/10.1016/j.eswa.2022.119446

[4] Eke C., Norman A., Shuib L., and Nweke H., “A Survey of User Profiling: State -of-the -Art, Challenges, and Solutions,” IEEE Access , vol. 7, pp. 144907 -144924, 2019. https://doi.org/10.1109/ACCESS.2019.2944243

[5] Feng W., Li Y., and Ma C., “Examining Non - English Foreign Language Education Through Social Media: Discourse and Psychological Analysis Based on Text Mining,” IEEE Access , vol. 12, pp. 152568 -152578, 2024. https://doi.org/10.1109/access.2024.3481049

[6] Koutsoumpis A., Oostrom J., Holtrop D., Van Breda W., and et al., “The Kernel of Truth in Text - Based Personality Assessment: A Meta -Analysis of the Relations Between the Big Five and the Linguistic Inquiry and Word Count (LIWC),” Psychological Bulletin , vol. 148, no. 11 -12, pp. 843 -868, 2022. https://doi.org/10.1037/bul0000381.supp

[7] Li Q., Cheng M., Wang J., and Sun B., “LSTM Based Phishing Detection for Big Email Data,” IEEE Transactions on Big Data , vol. 8, no. 1, pp. 278 -288, 2022. https://doi.org/10.1109/TBDATA.2020.2978915

[8] Li S. and Tang Y., “A Simple Framework of Smart Geriatric Nursing Considering Health Big Data and User Profile,” Computational and Mathematical Methods in Medicine , vol. 2020, no. 1, pp. 1 -9, 2020. https://doi.org/10.1155/2020/5013249

[9] Lin W., Xu H., Li J., Wu Z., and et al., “Deep - Profiling: A Deep Neural Network Model for Scholarly Web User Profiling,” Cluster Computing , vol. 26, no. 3, pp. 1753 -1766, 2023. https://doi.org/10.1007/s10586 -021 -03315 -2

[10] Liu X., Zhu Y., and Wu X., “Joint User Profiling with Hierarchical Attention Networks,” Frontiers of Computer Science , vol. 17, no. 3, pp. 1 -11, 2023. https://link.springer.com/article/10.1007/s11704 - 022 -1437 -6

[11] Maree M., Eleyat M., and Mesqali E., “Optimizing Machine Learning -Based Sentiment Analysis Accuracy in Bilingual Sentences via Preprocessing Techniques,” The International Arab Journal of Information Technology , vol. 21, no. 2, pp. 257 -270, 2024. https://doi.org/10.34028/iajit/21/2/8

[12] McHaney R., Tako A., and Robinson S., “Using LIWC to Choose Simulation Approaches: A Feasibility Study,” Decision Support Systems , vol. 111, pp. 1 -12, 2018. https://doi.org/10.1016/j.dss.2018.04.002

[13] Ouaftouh S., Zellou A., and Idri A., “Social Recommendation: A User Profile Clustering - Based Approach,” Concurrency and Computation -Practice and Experience , vol. 31, no. 20, pp. 1 -11, 2019. https://doi.org/10.1002/cpe.5330

[14] Pan Y., Huo Y., Tang J., Zeng Y., and Chen B., “Exploiting Relational Tag Expansion for Dynamic User Profile in a Tag -Aware Ranking Recommender System,” Information Sciences , vol. 545, pp. 448 -464, 2021. https://doi.org/10.1016/J.INS.2020.09.001

[15] Pflügner K., Maier C., Mattke J., and Weitzel T., “Personality Profiles that Put Users at Risk of Perceiving Technostress: A Qualitative Comparative Analysis with the Big Five Personality Traits,” Business and Information Systems Engineering , vol. 63, no. 4, pp. 389 -402, 2021. https://doi.org/10.1007/s12599 -020 -00668 -7

[16] Qian Y., Shao J., Zhang Z., Leng H., and et al., “Bert -CK: A Study of User Profile Classification Based on Bert and CK -Means Plus Fusion,” Journal of Intelligent and Fuzzy Systems , vol. 45, no. 3, pp. 4585 -4597, 2023. https://doi.org/10.3233/JIFS -224531

[17] Sharma S. and Gupta V., “Role of Twitter User Profile Features in Retweet Prediction for Big Data Streams,” Multimedia Tools and Applications , vol. 81, no. 19, pp. 27309 -27338, 2022. https://doi.org/10.1007/s11042 -022 -12815 -1

[18] Wang D., Wang P., Fu Y., Liu K., and et al., “Reinforced Imitative Graph Learning for Mobile User Profiling,” IEEE Transactions on Knowledge and Data Engineering , vol. 35, no. 12, pp. 12944 -12957, 2023. https://doi.org/10.1109/TKDE.2023.3270238

[19] Wang H., Tu Z., Fu Y., Wang Z., and Xu X., “Time -Aware User Profiling from Personal Service Ecosystem,” Neural Computing and Applications , vol. 33, no. 8, pp. 3597 -3619, 2021. https://doi.org/10.1007/s00521 -020 -05215 -9

[20] Widiyaningtyas T., Hidayah I., and Adji T., “User Profile Correlation -Based Similarity (UPCSim) Algorithm in Movie Recommendation System,” Journal of Big Data , vol. 8, no. 1, pp. 1 -21, 2021. https://doi.org/10.1186/s40537 -021 -00425 -x

[21] Yu Y., Li Q., and Liu X., “Automatic Anxiety Recognition Method Based on Microblog Text Analysis,” Frontiers in Public Health , vol. 11, pp. 1-6, 2023. https://doi.org/10.3389/fpubh.2023.1080013

[22] Zhang J., Han H., Su H., Jiang Z., and Peng C., “PUMTD: Privacy -Preserving User -Profile Matching Protocol in Social Networks,” China Communications , vol. 19, no. 6, pp. 77 -90, 2022. https://ieeexplore.ieee.org/document/9810136

[23] Zhao S., Li S., Ramos J., Luo Z., and et al., “User Profiling from their Use of Smartphone Applications: A Survey,” Pervasive and Mobile Computing , vol. 59, pp. 1 -20, 2019. https://doi.org/10.1016/j.pmcj.2019.101052