Article: Research on the application of personalised recommendation algorithm for university translation teaching based on big data analysis summary Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.81 pp.22 - 47 Abstract: This paper proposes a personalised recommendation algorithm based on big data analysis. The algorithm integrates multi-dimensional student information and optimises collaborative filtering methods to achieve efficient matching of translation learning resources. Firstly, a data collection system is established to gather student behaviour, achievements, resources, and learning environment data. Secondly, a dynamic analysis model is developed to integrate learner characteristics and cognitive level information. Finally, a translation opportunity comparison function is added to enhance resource accuracy and address data scarcity issues. The mean absolute error achieved 0.152, with the cumulative gain in normalised loss for the top 10 recommendations reaching 0.834. Practical application yielded significant outcomes: following 60 students' use of the recommendation system, average translation scores rose from 65.3 to 82.7 points, representing a 26.6% increase. Training engagement markedly improved, with session duration extending by 71.2% and interaction rates surging by 151.6%. Inderscience Publishers - linking academia, business and industry through research

Title: Research on the application of personalised recommendation algorithm for university translation teaching based on big data analysis summary

Authors: Xiaochun Lu; Fei Li

Addresses: School of Foreign Languages, Hubei Polytechnic University, Huangshi, 435000, Hubei, China ' Wuhan Vocational College of Software and Engineering (Wuhan Open University), School of General Education, Wuhan, 430205, Hubei, China

Abstract: This paper proposes a personalised recommendation algorithm based on big data analysis. The algorithm integrates multi-dimensional student information and optimises collaborative filtering methods to achieve efficient matching of translation learning resources. Firstly, a data collection system is established to gather student behaviour, achievements, resources, and learning environment data. Secondly, a dynamic analysis model is developed to integrate learner characteristics and cognitive level information. Finally, a translation opportunity comparison function is added to enhance resource accuracy and address data scarcity issues. The mean absolute error achieved 0.152, with the cumulative gain in normalised loss for the top 10 recommendations reaching 0.834. Practical application yielded significant outcomes: following 60 students' use of the recommendation system, average translation scores rose from 65.3 to 82.7 points, representing a 26.6% increase. Training engagement markedly improved, with session duration extending by 71.2% and interaction rates surging by 151.6%.

Keywords: big data analysis; translation teaching; personalised recommendation; collaborative filtering; learner portrait; education big data.

DOI: 10.1504/IJICT.2026.155114

International Journal of Information and Communication Technology, 2026 Vol.27 No.81, pp.22 - 47

Received: 10 Feb 2026
Accepted: 26 Mar 2026
Published online: 27 Jul 2026 *

Title: Research on the application of personalised recommendation algorithm for university translation teaching based on big data analysis summary

Keep up-to-date