User Task-Aware Re-ranking for Enhanced Web Search
DOI:
https://doi.org/10.36602/ijeit.v14i1.580Keywords:
User, Task, Web, Search, HCI, AIAbstract
This paper investigates task-aware re-ranking as a means of improving the relevance, efficiency, and effectiveness of web search systems. Traditional search engines often apply uniform ranking strategies without considering user intent, which limits their ability to align results with diverse search goals. To address this gap, we propose a task-centered framework that classifies queries into informational, navigational, and transactional categories, and re-ranks retrieved results accordingly. A fine-tuned GigaBERT model trained on the ORCAS-I dataset was employed to perform query classification, and a Task-Based Results Sorting System (TBRSS) was developed to evaluate the approach. A user study involving 33 participants compared baseline Google rankings with re-ranked outputs. Findings suggest that although statistical significance was limited, users perceived the task-aware system as more relevant, smoother, and faster. This research demonstrates the potential of integrating task-awareness into ranking algorithms to improve user experience, and highlights future directions including adaptive ranking, multilingual support, and larger-scale evaluations.
Downloads
References
[1] E. Agichtein, E. Brill, and S. Dumais, “Improving web search ranking by incorporating user behavior information,” in Proc. 29th Annu. Int. ACM SIGIR Conf. Res. Develop. Inf. Retrieval, 2006, pp. 19–26.
[2] R. Baeza-Yates, C. Hurtado, and M. Mendoza, “Improving search engines by query clustering,” J. Am. Soc. Inf. Sci. Technol., vol. 58, no. 12, pp. 1793–1804, 2007.
[3] G. W. You and S. W. Hwang, “Personalized ranking: A contextual ranking approach,” in Proc. ACM Symp. Appl. Comput., 2007, pp. 506–510.
[4] R. Faiz, “Analyzing temporal query for improving web search,” J. Emerg. Technol. Web Intell., vol. 4, no. 3, pp. 1–7, 2012.
[5] W. U. Ahmad, K. W. Chang, and H. Wang, “Context attentive document ranking and query suggestion,” in Proc. 42nd Int. ACM SIGIR Conf. Res. Develop. Inf. Retrieval, 2019, pp. 385–394.
[6] T. Guo et al., “Query-dominant user interest network for large-scale search ranking,” in Proc. Int. Conf. Inf. Knowl. Manage., 2023, pp. 629–638.
[7] B. J. Jansen and D. Booth, “Classifying web queries by topic and user intent,” in Proc. Conf. Hum. Factors Comput. Syst., 2010, pp. 4285–4290.
[8] B. MacKay and C. Watters, “Providing support for multi-session web tasks,” in Proc. ASIST Annu. Meeting, vol. 46, 2009.
[9] R. Xu, Y. Feng, and H. Chen, “ChatGPT vs. Google: A comparative study of search performance and user experience,” 2023. [Online]. Available: https://www.prolific.co/
[10] F. Ye, M. Fang, S. Li, and E. Yilmaz, “Enhancing conversational search: Large language model-aided informative query rewriting,” 2023. [Online]. Available: http://arxiv.org/abs/2310.09716
[11] C. Ziakis, M. Vlachopoulou, T. Kyrkoudis, and M. Karagkiozidou, “Important factors for improving Google search rank,” Future Internet, vol. 11, no. 2, pp. 1–15, 2019.
[12] A. Veglis and D. Giomelakis, “Search engine optimization,” Future Internet, vol. 12, no. 1, pp. 1–15, 2020.
[13] A. Giannakoulopoulos, N. Konstantinou, D. Koutsompolis, M. Pergantis, and I. Varlamis, “Academic excellence, website quality, SEO performance: Is there a correlation?” Future Internet, vol. 11, no. 11, pp. 1–18, 2019.
[14] A. Alhenshiri, C. Watters, and M. Shepherd, “User behaviour during web search as part of information gathering,” in Proc. Annu. Hawaii Int. Conf. Syst. Sci., 2011.
[15] Y. Xi et al., “Multi-level interaction reranking with user behavior history,” in Proc. 45th Int. ACM SIGIR Conf. Res. Develop. Inf. Retrieval, 2022, pp. 1336–1346.
[16] P. Surana, P. Jamdade, T. Nandedkar, A. S. Ghadge, and N. Bobde, “Enhancing the search result for user query using iterative user feedback,” Int. J. Innov. Res. Comput. Commun. Eng., vol. 3, no. 1, pp. 71–76, 2015.
Downloads
Published
License
Copyright (c) 2025 The International Journal of Engineering & Information Technology (IJEIT)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
