روشی برای شناسایی تاثیر اجتماعی با استفاده از شباهت تعبیه شده و هوموفیلی در شبکه های اجتماعی مکان مبنا
محورهای موضوعی : مهندسی کامپیوتر و فناوری اطلاعات
زهره سادات اخوان حجازی
1
,
مهدی اسماعیلی
2
,
مصطفی قبایی آرانی
3
,
بهروز مینایی بیدگلی
4
1 - گروه مهندسی کامپیوتر، واحد قم، دانشگاه آزاد اسلامی، قم، ایران
2 - گروه مهندسی کامپیوتر، واحد کاشان، دانشگاه آزاد اسلامی، کاشان، ایران
3 - گروه مهندسی کامپیوتر، واحد قم، دانشگاه آزاد اسلامی، قم، ایران
4 - گروه مهندسی کامپیوتر، دانشگاه علم و صنعت ایران، تهران، ایران
کلید واژه: تاثیر اجتماعی, کاربران اجتماعی, شباهت تعبیه شده, هوموفیلی, شبکه¬های اجتماعی مکان مبنا.,
چکیده مقاله :
شبکههای اجتماعی مکانمبنا به دلیل دادههای ارزشمند موقعیت مکانی و روابط کاربران، در حوزه تجارت و تبلیغات هدفمند بسیار مهم هستند. یک موضوع کلیدی در این شبکهها، تعیین نفوذ اجتماعی کاربران برای شناسایی افراد تأثیرگذار است. تحقیقات اخیر، نقش شباهت گرهها را در تعیین تأثیر اجتماعی کاربران به روشنی نشان داده است. از بین روشهای محاسبه شباهت گرهها، بسیاری از روشهای موجود جنبههای متعدد شباهت بین کاربران را نادیده میگیرند. در این مقاله، با ترکیب دو معیار هوموفیلی و شباهتهای تعبیهشده و پایه، تأثیر اجتماعی کاربران درشبکههای مکانمبنا بررسی شده است. نتایج نشان میدهد شباهت ترکیبی بهویژه در مواقعی که کاربران دارای ویژگیهای مشابه هستند، موجب افزایش دقت در شناسایی کاربران با نفوذ بیشتر میشود. این بدان معناست که ویژگیهای شخصی و اجتماعی کاربران، علاوه بر روابط ساختاری آنها در شبکه، نقش تعیینکنندهای در تحلیل رفتارهای اجتماعی ایفا میکنند. استفاده از مدلهای تعبیه گره باعث بهبود دقت شباهتهای ساختاری و تشخیص روابط پیچیده میان گرهها شد. با بکارگیری روش پیشنهادی برروی مجموعه دادههای واقعی، نشان داده شد که استفاده از مدلهای تعبیه گره بدلیل سادگی و عدم صرف زمان برای آموزش دادهها، در پردازش دادههای پیچیدهتر و بزرگتر درمقایسه با روشهای پیشرفته، از نظر محاسباتی و هزینه قابل رقابت میباشند.
Location-based social networks are very important in the field of targeted commerce and advertising due to their valuable location data and user relationships. A key issue in these networks is determining the social influence of users to identify influential users. Research has proven the role of node similarity in finding the social influence of users. Among the methods for calculating node similarity, many existing methods ignore multiple aspects of similarity between users. In this paper, by combining two criteria of homophily and embedded and basic similarities, the social influence of users in location-based networks is investigated. The results show that combined similarity increases the accuracy in identifying more influential users, especially when users have similar characteristics. This means that the personal and social characteristics of users, in addition to their structural relationships in the network, play a decisive role in analyzing social behaviors. The use of node embedding models improved the accuracy of structural similarities and the detection of complex relationships between nodes. By applying the proposed method on real datasets, it was shown that the use of node embedding models, due to their simplicity and lack of time spent on data training, is computationally and cost-competitive in processing more complex and larger data compared to advanced methods.
[1] C. C. Aggarwal, An Introduction to Social Network Data Analytics, Springer US, 2011, doi: 10.1007/978-1-4419-8462-3_1.
[2] R. Panchendrarajan and A. Saxena, "Topic-based influential user detection: a survey," Applied Intelligence, vol. 53, no. 5, pp. 5998-6024, 2023, doi: 10.1007/s10489-022-03831-7.
[3] R. Rabade, N. Mishra, and S. Sharma, "Survey of influential user identification techniques in online social networks," in Recent Advances in Intelligent Informatics: Proceedings of the Second International Symposium on Intelligent Informatics (ISI'13), Mysore, India, Aug. 2013, pp. 359-370, Springer, 2014, doi: 10.1007/978-3-319-01778-5_37.
[4] S. Peng, Y. Zhou, L. Cao, S. Yu, J. Niu, and W. Jia, "Influence analysis in social networks: A survey," Journal of Network and Computer Applications, vol. 106, pp. 17-32, 2018, doi: 10.1016/j.jnca.2018.01.005.
[5] X. Wei, Y. Qian, C. Sun, J. Sun, and Y. Liu, "A survey of location-based social networks: problems, methods, and future research directions," GeoInformatica, vol. 26, no. 1, pp. 159-199, 2022, doi: 10.1007/s10707-021-00450-1.
[6] S. Khetarpaul, "Mining location based social networks to understand the citizen’s check-in patterns," Computing, vol. 103, no. 12, pp. 2967-2993, 2021, doi: 10.1007/ 10.1007/s00607-021-01020-x.
[7] X. Yang, "Identification of influential spreaders in geo-social network," in Proc. 25th Int. Conf. on Geoinformatics, Aug. 2017, pp. 1-4, doi: 10.1109/GEOINFORMATICS.2017.8090941.
[8] V. Iswarya, V. Govindasamy, and V. Akila, "A survey on the identification of influential spreaders in complex networks," in Proc. 2nd Int. Conf. on Computer, Communication and Control (IC4), Feb. 2024, pp. 1-4, doi: 10.1109/IC457434.2024.10486364.
[9] X. Pan, R. Hu, and D. Li, "Social-IFD: Personalized influential friends discovery based on semantics in LBSN," in Proc. IEEE Int. Conf. on Communications (ICC), Jun. 2020, pp. 1-6, doi: 10.1109/ICC40277.2020.9148703.
[10] K. Ali, C. T. Li, and Y. S. Chen, "Joint selection of influential users and locations under target region in location-based social networks," Sensors, vol. 21, no. 3, p. 709, 2021, doi: 10.3390/s21030709.
[11] K. Zhang and K. Pelechrinis, "Understanding spatial homophily: the case of peer influence and social selection," in Proc. 23rd Int. Conf. on World Wide Web, Apr. 2014, pp. 271-282, doi: 10.1145/2566486.2567990.
[12] Z. S. Akhavan-Hejazi, M. Esmaeili, M. Ghobaei-Arani, and B. Minaei-Bidgoli, "Identifying influential users using homophily-based approach in location-based social networks," The Journal of Supercomputing, pp. 1-36, 2024, doi: 10.1007/s11227-024-06228-0.
[13] A. Zareie and R. Sakellariou, "Similarity-based link prediction in social networks using latent relationships between the users," Scientific Reports, vol. 10, no. 1, p. 20137, 2020, doi: 10.1038/s41598-20-76799-4.
[14] X. Chen, L. Deng, Y. Zhao, X. Zhou, and K. Zheng, "Community-based influence maximization in location-based social network," World Wide Web, vol. 24, pp. 1903-1928, 2021, doi: 10.1007/s11280-021-00935-x.
[15] N. U. Khan, W. Wan, R. Riaz, S. Jiang, and X. Wang, "Prediction and classification of user activities using machine learning models from location-based social network data," Applied Sciences, vol. 13, no. 6, p. 3517, 2023, doi: 10.3390/app13063517.
[16] J. Zhao, Y. Song, F. Liu, and Y. Deng, "The identification of influential nodes based on structure similarity," Connection Science, vol. 33, no. 2, pp. 201-218, 2021, doi: 10.1080/09540091.2020.1806203.
[17] A. M. U. D. Khanday, M. A. Wani, S. T. Rabani, and Q. R. Khan, "Hybrid approach for detecting propagandistic community and core node on social networks," Sustainability, vol. 15, no. 2, p. 1249, 2023, doi: 10.3390/su15021249.
[18] J. A. Smith, M. McPherson, and L. Smith-Lovin, "Social distance in the United States: Sex, race, religion, age, and education homophily among confidants, 1985 to 2004," American Sociological Review, vol. 79, no. 3, pp. 432-456, 2014, doi: 10.1177/0003122414531776.
[19] R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014, doi: 10.1017/CBO9781139088510.
[20] K. Z. Khanam, G. Srivastava, and V. Mago, "The homophily principle in social network analysis: A survey," Multimedia Tools and Applications, vol. 82, no. 6, pp. 8811-8854, 2023, doi: 10.1007/s11042-021-11857-1.
[21] B. Perozzi, R. Al-Rfou, and S. Skiena, "Deepwalk: Online learning of social representations," Proc. 20th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, 2014, pp. 701-710, doi: 10.1145/2623330.2623732.
[22] M. Grohe, "word2vec, node2vec, graph2vec, x2vec: Towards a theory of vector embeddings of structured data," Proc. 39th ACM SIGMOD-SIGACT-SIGAI Symp. on Principles of Database Systems, 2020, pp. 1-16, doi: 10.1145/3375395.3387641.
[23] P. Goyal and E. Ferrara, "Graph embedding techniques, applications, and performance: A survey," Knowledge-Based Systems, vol. 151, pp. 78-94, 2018, doi: 10.1016/j.knosys.2018.03.022.
[24] J. Zhou, L. Liu, W. Wei, and J. Fan, "Network representation learning: from preprocessing, feature extraction to node embedding," ACM Computing Surveys (CSUR), vol. 55, no. 2, pp. 1-35, 2022, doi: 10.1145/3491206.
[25] A. Grover and J. Leskovec, "node2vec: Scalable feature learning for networks," Proc. 22nd ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, 2016, pp. 855-864, doi: 10.1145/2939672.2939754.
[26] C. Zhang, T. Li, Y. Gou, and M. Yang, "KEAN: Knowledge embedded and attention-based network for POI recommendation," Proc. IEEE Int. Conf. on Artificial Intelligence and Computer Applications (ICAICA), 2020, pp. 847-852, doi: 10.1109/ICAICA50127.2020.91822385.
[27] A. Kosmatopoulos, K. Loumponias, D. Chatzakou, T. Tsikrika, S. Vrochidis, and I. Kompatsiaris, "Random-walk graph embeddings and the influence of edge weighting strategies in community detection tasks," Proc. 2021 Workshop on Open Challenges in Online Social Networks, 2021, pp. 9-13, doi: 10.1145/3472720.3483621.
[28] A. Saxena, G. Fletcher, and M. Pechenizkiy, "Nodesim: node similarity-based network embedding for diverse link prediction," EPJ Data Science, vol. 11, no. 1, p. 24, 2022, doi: 10.1140/epjds/s13688-022-00336-8.
[29] T. Tao, Q. Wang, Y. Ruan, X. Li, and X. Wang, "Graph Embedding with Similarity Metric Learning," Symmetry, vol. 15, no. 8, p. 1618, 2023, doi: 10.3390/sym15081618.
[30] D. Chen, P. Du, B. Fang, D. Wang, and X. Huang, "A node embedding-based influential spreaders identification approach," Mathematics, vol. 8, no. 9, p. 1554, 2020, doi: 10.3390/math8091554.
[31] D. Yang, B. Qu, J. Yang, and P. Cudre-Mauroux, "Revisiting user mobility and social relationships in LBSN: a hypergraph embedding approach," The Web Conference (WWW), 2019, pp. 2147-2157, doi: 10.1145/3308558.3313635.
[32] D. Yang, B. Qu, J. Yang, and P. Cudré-Mauroux, "Lbsn2vec++: Heterogeneous hypergraph embedding for location-based social networks," IEEE Trans. on Knowledge and Data Engineering, vol. 34, no. 4, pp. 1843-1855, 2022, doi: 10.1109/TKDE.2020.2997869.
[33] W. Liu, S. Wang, and J. Ding, "Influence Maximization Based on Adaptive Graph Convolution Neural Network in Social Networks," Electronics, vol. 13, no. 16, p. 3110, 2024, doi: 10.3390/electronics13163110.
[34] Z. Liu, P. Yang, Q. Hao, W. Zheng, and Y. Xiao, "Exploring implicit influence for social recommendation based on GNN," Soft Computing, pp. 1-14, 2024, doi: 10.1007/s00500-024-09898-3.
[35] H. Zhang, G. Kou, Y. Peng, and B. Zhang, "Role-aware random walk for network embedding," Information Sciences, vol. 652, p. 119765, 2024, doi: 10.1016/j.ins.2023.119765.
[36] J. Liu, H. Yi, Y. Gao, and R. Jing, "Personalized Point-of-Interest Recommendation Using Improved Graph Convolutional Network in Location-Based Social Network," Electronics, vol. 12, no. 16, p. 3495, 2023, doi: 10.3390/electronics12163495.
[37] Y. Shang, B. Zhou, X. Zeng, Y. Wang, H. Yu, and Z. Zhang, "Predicting the popularity of online content by modeling the social influence and homophily features," Frontiers in Physics, vol. 10, p. 915756, 2022, doi: 10.3389/fphy.2022.915756.
[38] Stanford SNAP, "Gowalla Dataset," [Online]. Available: https://snap.stanford.edu/data/loc-gowalla.html. Accessed: Jan. 14, 2016.
[39] Stanford SNAP, "Brightkite Dataset," [Online]. Available: https://snap.stanford.edu/data/loc-brightkite.html. Accessed: Jan. 14, 2016.
[40] J. J. Levandoski, M. Sarwat, A. Eldawy, and M. F. Mokbel, "Lars: A location-aware recommender system," Proc. IEEE 28th Int. Conf. on Data Engineering (ICDE), 2012, pp. 450-461, doi: 10.1109/ICDE.2012.54.
