Aniruddha Chattopadhyay

Competetions

• Winner HSBC Hackathon on AI , IIT Kharagpur

  • Built an intelligent medical chat-bot using tkinter(GUI) in python.
  • Chat bot takes in the symptom of the user as input and processes that using Natural Language Processing.
  • The processed input is then put through a Multi layer perceptron Neural network that maps it to a keyword(Symptom).
  • The Symptom is then searched in NEO4J graph database and other relevant symptoms the user may have is asked as cross query to the user.
  • Finally using the number of symptoms matched of a particular disease , a conclusive health report is given to the user.
  
  Here is the certificate for the competition
  



• First Runners up of National AI Hackathon with theme 'AI for society' in VESAITHON, MUMBAI

  • The theme of the hackathon was to build AI solutions for social challenges faced by senior citizens, differently abled and kids. The major sponsors were Capgemini, Network Marvels, Adapty and NVIDIA.
  • Built two android applications (in Android Studios) one for the elderly people and one for the caregivers.
  • The app for the elderly people had a neural net trained on the MOBIACT dataset stored as a tensorflow lite model in the phone to detect falls and alert the caregiver.
  • Used AWS Rekognize API to detect the emotion of the person and keep a record of the emotions with a timestamp in the application of the caregiver.
  • Used IBM Watson to create a personalized multilingual voice controlled chatbot for the elderly person. The chatbot is automatically updated with the mood of the person and accordingly fine tunes the conversation with the elderly person. If the person in sad ,it can automatically play funny videos or a nostalgic song to cheer up the person. It can also look up for doctors nearby, search for hospitals, give a call to any of the contacts including the caregiver.
  • The app keeps track of medical history, allergy history as well as give medicine reminder to both the elderly as well as the caregiver. The medicine to be taken can be added as a image and the app will automatically update its database in FireBase with the name of the medicine using text recognition in images.
  • Real time Location tracking as well as a pedometer is built in as features of the app.
  • The app got huge media coverage from TOI,India Today, Anandabazar patrika.
  
  Here is the certificate for the competition
  



• Top 8 finalist in National Transportation Hackathon by NEC Corp, DELHI

  Made an android app to make Indian transportation more Predictable
  • The app allows users to do real-time tracking of bus/autorickshaw and also gives them estimate time of waiting at the bus stop along with estimated journey time so that they can optimize their journey schedules accordingly.
  • It also gives them the option of choosing from either cheapest or fastest means of transport available. Gives them all possible combinations of traveling options
  • The app also shows the users how crowded a bus is and accordingly lets them choose whether they want to travel via this bus or wait or the next one. The dilemma that users face is if they leave a bus just right now they don’t know how long will they have to wait for the next bus but our app addresses that by showing all the bus that is going to pass through that route and with an estimated time of arrival, estimate time of journey and an estimated crowd. Thus making the whole journey predictable.
  
  Here is the certificate for the competition
  



• Smart India Hackathon 2020 Finalist

  We reached the finals of the Smart India Hackathon 2020.
  
  Find the link for the demo Video and presentation