Youssof Kamal | Abdelhady Eslam | Mahmoud Hamed | Abdelrahman Naser | Abdelrahman Salah | Mohamed Sayed
People sometimes need recommendations or impressions about someone specific products, trends and more. They won't waste a lot of time reading all opinions
From this point of view and to overcome this problem, we have built a web app using some data mining algorithms that helps analyzing and exposing content of review or blog post written about some product or trend that helps someone deciding if that content suitable for him or not
The platform holds these features:
- Making profile
- Follow profiles system
- Writing posts
- Like/Dislike and commenting
- Notifications
- Unique News Feedback for each user
- Full-Text search (words/tags)
We have used Prototype Model. The basic idea in Prototype model is that: Prototype model is a software development model that's developed based on the currently known requirements. A throwaway prototype is built to understand the requirements. By using this prototype, the client can get an "actual feel" of the system, since the interactions with prototype can enable client to better understand the requirements of the desired system. Prototyping is an attractive idea for complicated and large systems for which there is no manual process or existing system to help determine the requirements. Prototypes are usually not complete systems and many of the details are not builtin the prototype. The goal is to provide a system with overall functionality
Tools and technologies
-
DataBases
MySQL - MongoDB -
BackEnd
Django -
FrontEnd
HTML - CSS - Bootstrap - JS - JQuery -
Classifications Algorithms
Logistic Regression - KNN - SVC - Decision Tree - Random Forests -
Python Packages
re - Pandas - Stopwords from nltk.corpus - word_tokenize from nltk.tokenize - PorterStemmer from nltk.stem.porter
- News Feed Generation (PageRank Algorithm -social networks-)