Flipkart's Project Mira: Your own personalised shopping assistant using Artificial Intelligence

India’s largest ecommerce player Flipkart is working on an artificial intelligence focused approach to provide its customers with an offline shopping experience when they shop online. According to a report by the Economic Times, Flipkart has soft launched Project Mira, an artificial intelligence-focused effort last February.

What started out as CEO Binny Bansal’s vision of using new technology to transform how Indians shop online, Flipkart has been silently running experiments on its app and website to understand customers better.

The idea here is to help buyer search for what they want, get personalized recommendations, talk to shoppers while they search for products. Simply put, Flipkart is looking to replicate the offline shopping experience online.

Speaking with ET, Ram Papatla, vice president of product, Flipkart said, “On 28 February (2017), we launched the first version of our conversational search experience. Now, our users with broad intent (searching for, say, shoes or bedsheets) are guided with relevant questions, conversational filters, shopping ideas, offers and trending collections.”

This something that has already been experimented on extensively by global ecommerce majors like eBay and Amazon.

The advantage Flipkart has in India is the amount of user data to analyze and effectively make shopping experiences easy for customers.

“When we looked at the (product) returns data and when we looked at data from shoes and lifestyle (categories), we saw a bunch of mismatch of expectations from our customers in terms of size and fit issues. If only we could have asked them one question we could have given the right response,” Papatla told ET. He adds that Flipkart has enough evidence (to say) that had it ‘talked’ to Mira, the problem could have been solved.

For every four products purchased on Flipkart, at least one was being returned, especially in categories like clothing and accessories.

Flipkart says it expects AI-led innovation through Project Mira will help it create solid competitive differentiation and cement its market leadership

According to the ET report, the machine learning algorithms can also detect incorrect images and morphed images. Posting the right images on the platform is crucial to getting customers interested in a product. “Another pressing problem is duplicate products. Sellers intentionally or unintentionally post duplicate products, which increases user effort in scanning to the desired set of products. Finally, we want to ensure that the descriptors of a product like colour, pattern, etc., are accurate,” said Papatla.

Flipkart is also parterning with several academic institutions like IIT-Kharagpur/Kanpur/Bombay, Carnegie Mellon University and IIIT-Hyderabad for research on various new technologies.
ET reports that IIT-Kharagpur is also working on a chatbot that can answer product queries from Flipkart’s customers. “We have built the initial prototype using sequence-tosequence deep-learning framework that, given a customer query, provides up to five possible responses,” said Pawan Goyal, assistant professor at the Department of Computer Science. “During the initial evaluation, the correct response was found within these top five responses returned by the system.”

Even with Project Mira only in an infancy stage, Flipkart is seeing nearly a 12% cart addition and a dramatic improvement of seller experience.

It is now working on extending Mira to resolve issues like product returns and quicker delivery. “How do you employ machine learning models to see if a delivery will lead to returns and for what reasons? We are also trying to estimate if we can deliver, say, in (an optimal) two days, because if it takes longer customers fall off, they cancel midway,” said Mayur Datar, chief data scientist at Flipkart. 

Image: Flipkart Facebook Page