VertiKin
VertiKin is an e-commerce platform which allows the user to search through various products. It has a one-of-its-kind feature of automatically detecting what user might be interested in buying in near future.
How VertiKin works
VertiKin Mobile app learns from user inputs on his phone (we do not read passwords and private information, so the user can be assured of his security). This data is sent to the VertiKin server, where using Natural language processing (NLP), important information is extracted and using frequency as a parameter, it tries to find if the user is interested in buying some product or not. If so, a GCM push notification is sent to the user.
VertiKin Improves itself
Let's say VertiKin realised user wants to buy product A, but user did not want it, we give him a small feedback form, where he simply needs to click No and VertiKin would improve the suggestion from next time. User starts with a DEFAULT_THRESHOLD and according to the user feedback, user threshold changes by THRESHOLD_DELTA.
Impact
- According to a 2010 study by The Nielsen Company, people tend to talk more about their needs and reviews influencing other people's opinion. So we used this information to detect, if a person tends to buy something, he tends to talk about it.
- Cognitive fluency is the human tendency to prefer things that are familiar and easy to understand. According to this article published on boston.com, people buy stuff which is easier to understand. So using this information and the fact that after a certain point of time our primary source of knowledge becomes our peer group, we can deduce it is highly likely a person discusses a product with his friends / family.
Screenshot

External Tools / APIs used
- Custom Keyboard by Mehdi Mirzaei
- nltk
- Google Cloud Messaging
- Walmart Open API


Log in or sign up for Devpost to join the conversation.