Precision trichology: A patient-feedback driven framework for personalized haircare recommendations

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

The Haircare Recommendation System is an exceptional way to receive tailored suggestions on haircare products based on the constituents of hair products and personal hair concerns. Using machine learning, and, in particular, Natural Language Processing (NLP) and a recommendation system, the system identifies the users' data collected through a survey, against a curated and expert validated haircare product dataset. The system then can recommend products and/or ingredients that are the most helpful based on users' hair concerns, such as dryness, dandruff, hair loss, etc. Choosing haircare products is made easier through this product by decanting complex ingredient lists and providing science-laced suggestions to each user. This system provides users with the ability to purchase products that are beneficial and helpful to their hair. The system also is implemented to be an all-inclusive product that collects users' suggestions and is made to be modern and up-to-date at all times. Using a TF-IDF-based logistic regression model on Amazon product reviews, the classification of reviews with respect to the positive (≥4.0 stars) and the negative reviews achieved an accuracy between 85–95%.