Recom-D

Introduction of the Application.
Selection of Medical Conditions to get Drugs Recommendation.
Suggestions of Drugs as per the Medical Condition.
Details of particular Drug.
Github Repository of Recom-D.

Inspiration

There are numerous medications available for sickness, and sometimes the consumers may have difficulty in selecting the best prescriptions for their illnesses. So Recom-D is created to help to elucidate this matter such that it provides the users with multiple drugs as recommendations with their detailed information like Dosage Form, Route, Strength of the drug to identify and select the best drugs for their medications.

What it does

Recom-D is an AI based Medicine recommendation System that Recommends the user with various drugs considering the customer review/sentiments and their rating towards the drugs in the past. It uses various features like Rating, Reviews etc to achieve this feat. While suggesting the drugs for medication it also provides in-depth details of the drugs for the users to give them clear understanding for drugs.

How we built it

The Recom-D is built using various tool and technologies. Firstly the data was collected from UCI Machine Learning Repository. Later the Data was cleaned and processed with the NLTK for better analysis. The graphs are generated using Matplotlib and Seaborn to find Insights and provide easy understanding of dataset. The Customer Ratings and Reviews were considered as most important features to target the drugs that could be suggested. Lastly we have used Vercel for deployment of the project where frontend is done using React and for backend Flask is been used.

Challenges we ran into

There were many obstacles and challenges that we ran into during creation of the Recom-D. The few of them were:-

Deployment of the AI Model was the most Crucial and Important task as it provides easy interpretation and Understanding of the Drugs.
Cleaning and Getting insights of the Data to Recommend the Drugs.
Performing EDA to find better Insights.
Finding the best appropriate dataset for analysis.

Accomplishments that we're proud of

We are proud of the model's deployment since, despite several challenges, we achieved the model's deployment with great accuracy. This project will be easy to interpret and handy for receiving the finest medicine recommendations with detailed description at your fingertips.

What we learned

We learnt about various things like Team and Time Management, Individual Decision Making, Communications, Deployment etc. We are also obliged that we have also gained knowledge and interest about many technologies and tools like GitHub, Vercel, React, Flask, NLTK, Python, Pandas and Many More.

What's next for Recom-D

Recom-D has wide variety of future options available some like, Recom-D can be offered for a more variety of illnesses when a sufficient quantity of data has been gathered. It can also provide advancement like places to buy Drugs (Online/Offline Centres) with its navigations and Recom-D could also provide access to on-site medication procurement. Other than these it can be used to detect diseases using Computer Vision and suggest Drugs accordingly.