Optimizing E-commerce Evaluations with PSO (MACHINE LEARNING, PYTHON)
Budget: ₹600 – ₹1,500 INR
This project seeks to improve the accuracy of product evaluations on e-commerce platforms through innovative techniques. Despite not specifying the particular platforms, all applications will be considered, supporting Amazon, eBay, Shopify, and others. General evaluations will be improved, including product ratings, vendor ratings, and user reviews.
The broader aim of the project is to develop a robust system to handle the indeterminate and inconsistent user-generated content, primarily identified as inconsistent ratings, indeterminate feedback, and lack of optimization. To achieve this, we propose to utilize neutrosophic sets and Particle Swarm Optimization (PSO).
Ideal skills for this work include:
- Strong knowledge in neutrosophic sets and Particle Swarm Optimization
- Experience in data analytics and optimization techniques for user-generated content.
WHAT I DID
I cleaned and preprocessed the dataset
Thn i applied sentiment analysis combined with neutrosophic set for better accuracy and the membership function values are being calculated dynamically
Thn i aggregate the file created once by each product and thn for entire dataset
Thn i applied pso to find the optimized weights of t,i, f and for each product
WHAT I WANT
Train and test the model and accuracy, precision, also find f1 score and recall
The broader aim of the project is to develop a robust system to handle the indeterminate and inconsistent user-generated content, primarily identified as inconsistent ratings, indeterminate feedback, and lack of optimization. To achieve this, we propose to utilize neutrosophic sets and Particle Swarm Optimization (PSO).
Ideal skills for this work include:
- Strong knowledge in neutrosophic sets and Particle Swarm Optimization
- Experience in data analytics and optimization techniques for user-generated content.
WHAT I DID
I cleaned and preprocessed the dataset
Thn i applied sentiment analysis combined with neutrosophic set for better accuracy and the membership function values are being calculated dynamically
Thn i aggregate the file created once by each product and thn for entire dataset
Thn i applied pso to find the optimized weights of t,i, f and for each product
WHAT I WANT
Train and test the model and accuracy, precision, also find f1 score and recall