APPLICATION OF SOCIOPHYSICS CONCEPT IN ELECTRONIC BOOK RECOMMENDATION SYSTEM USING MACHINE LEARNING METHODS K-NEAREST NEIGHBOR (KNN) AND RANDOM FOREST
Physics as a field of study that explores natural phenomenas has undergone various developments, one of which is the emergence of complex systems as a field of study. Complex systems explain how interactions among the components of a system, whether in macro or micro conditions, dynamically influ...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/80602 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Physics as a field of study that explores natural phenomenas has undergone various
developments, one of which is the emergence of complex systems as a field of
study. Complex systems explain how interactions among the components of a
system, whether in macro or micro conditions, dynamically influence each other.
This resembles human interactions, including behaviors that affect such
interactions. Departing from it, the author is interested in examining the application
of physics concepts within socio-physics to understand the complexity of social
interactions. This Final Project aims to develop a recommendation system by
applying K-Nearest Neighbor and Random Forest methods to predict the ratings of
books that users have not yet read. The research results indicate that the KNN model
tends to produce relatively uniform and sequential book ratings, indicating potential
overfitting. In contrast, the RF model shows significant variation in the rankings of
registered books. Evaluation metrics, such as Mean Absolute Percentage Error
(MAPE), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and
Overall Accuracy (OA), are used to compare the performance of both methods.
Empirical findings show that RF (MAPE: 7.90%; MAE: 0.3; RMSE: 0.40; OA:
0.84) consistently outperforms KNN (MAPE: 16.74%; MAE: 0.6; RMSE: 0.68;
OA: 0.28)
|
|---|