Abstract

This paper explores the integration of machine learning (ML) applications in social media influencer marketing, using ASOS as a case study to examine its effectiveness in optimizing influencer-brand collaborations. With the digital marketing landscape rapidly evolving, traditional metrics like follower counts are no longer sufficient for effective influencer selection, particularly when engaging with Generation Z's preference for authenticity and positivity. Given Gen Z's significant role in influencer marketing, it's vital for influencers to align with brand values, reflecting this demographic's authentic preferences. Our study constructs a novel ML-based recommender system designed to enhance the precision of influencer selection by leveraging deep learning and natural language processing techniques. This system categorizes influencers, analyzes engagement rates, and ensures demographic alignment, tackling the scalability and accuracy challenges in influencer marketing. Through a cosine similarity model, we quantitatively evaluate the alignment between influencers and ASOS’s brand identity and goals, measuring the cosine of the angle between feature vectors of both entities. Our analysis, leveraging ASOS's API data and influencers' Instagram metadata, reveals significant improvements. The system achieved an F1 score of 0.74, indicating a balanced enhancement in precision and recall rates compared to a baseline model performance with an F1 score of 0.48. This result underscores the system's efficacy in identifying influencers who resonate with Generation Z, presenting a substantial advancement over traditional selection methods.

Advisor

Musgrave, John

Department

Computer Science

Publication Date

2024

Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis

Download

Share

COinS
 

© Copyright 2024 Ephrathah Gebremichael