Skip to content

Analyzed a dataset comprising over 1,000 products to reveal consumer preferences and identify e-commerce trends. Employed rigorous data preparation and analysis techniques to delve into the extensive product landscape on Amazon. Developed a recommendation system and performed sentiment analysis using machine learning.

Notifications You must be signed in to change notification settings

wonderakwei/Product-Performance-Analysis-and-Recommendation-System

Repository files navigation

Product-Performance-Amazon-Product-Analysis-and-Recommendation-System

Amazon Sales Dataset Project

Overview

The Amazon Sales Dataset was a detailed collection of over 1,000 Amazon product ratings and reviews, sourced directly from Amazon's official website. This dataset provided profound insights into consumer behaviors, product trends, and user sentiments.

Key Features

  • Product Details: Entries contained vital data like product ID, name, category, and pricing specifics.

  • Price Metrics: The dataset illuminated product pricing dynamics, showcasing both discounted and original prices along with discount percentages.

  • Rating Insights: Detailed the overall product ratings and the number of contributors to each rating.

  • User Feedback: Entries provided comprehensive user review data, including user ID, name, review titles, content, and associated review ID.

  • Product Media: Included direct links to product images and the respective Amazon product pages.

Inspiration

Amazon, a groundbreaking American tech giant, has reshaped the e-commerce paradigm. Covering everything from inventory to customer experience, Amazon stands out in the retail sector. This project successfully extracted insights, proposed innovative solutions, and tapped into the immense potential of e-commerce data.

The project's purpose was to decode consumer inclinations, interpret buying patterns, and craft a recommendation mechanism to align products with user preferences.

Project Outline

  1. Data Collection: The Amazon products dataset was obtained from Kaggle.

  2. Data Preparation: Extensive work was done to explore, clean, and preprocess the dataset, ensuring its readiness for thorough analysis.

  3. Exploratory Data Analysis (EDA): Deep dives into the dataset unveiled product distribution, customer ratings, and reviews across categories.

  4. Data Visualization: Various visual tools were employed to present data trends and patterns effectively.

  5. Sentiment Analysis: The VADER Sentiment Scoring method from the NLTK library was utilized to gauge sentiments from reviews, highlighting correlations with product ratings.

  6. Recommendation System: Leveraged machine learning to design a recommendation system, providing users with product suggestions tailored to their tastes and past interactions.

Sentiment Analysis

Using the VADER Sentiment Scoring method, sentiment insights were extracted from the reviews. The analysis distinctly showcased the relationship between product ratings and the sentiments conveyed in the reviews.

Recommendation System

The project aimed to elevate user experience by tailoring product suggestions. By harnessing the Amazon dataset, which encompassed product, review, and user details, a recommendation system was developed. This system's personalized recommendations enriched user interactions, strengthening business-consumer relationships and fostering revenue growth.

About

Analyzed a dataset comprising over 1,000 products to reveal consumer preferences and identify e-commerce trends. Employed rigorous data preparation and analysis techniques to delve into the extensive product landscape on Amazon. Developed a recommendation system and performed sentiment analysis using machine learning.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published