An Investigation into the Construction of Learner Portraits Grounded in Peer Influence

Abstract

In the context of the digital transformation of education, this study focuses on how to integrate the theory of peer effects to construct a learner profile model that combines individual characteristics with group interaction data, and to optimize teaching strategies using
this model. By calculating the stylistic and behavioral similarities among learners, and combining the cosine similarity and Hamming distance
formulas, a learner group characteristic model has been constructed. Meanwhile, indicators such as fractional ranking and learning efficiency
have been introduced to measure the influence relationships among learners, forming a stable learner group and evaluating the group stability
using the Dunn Index. This model is capable of accurately capturing the individual needs and preferences of learners, providing a basis for
personalized learning resource recommendations and effectively enhancing the matching degree and utilization efficiency of learning resources. Moreover, the model offers data support for teachers to optimize teaching strategies, facilitating the design of teaching activities such as
differentiated instruction and cooperative learning, and significantly improving teaching effectiveness. The research findings hold significant
importance for promoting educational equity and quality enhancement.