Amazon India is devising ways to up its market share using AI and machine learning
Rajeev Rastogi, director for machine learning at Amazon, said both transformative technologies are being used in several ways to improve customer experience. For instance, he said, apart from guiding sellers on categorisation, Amazon also tracks products that have high sales during festivities or holidays, and based on such factors, it suggests the best products and deals that can woo customers.
All of these products and services are being built in India, he said.
The Seattle-headquartered company soon plans to launch event-based deals in India. These deals will specifically be offered on occasions such as Rakshabandhan or Diwali. These deals are currently available in its US market.
“Advertising is another area where we use machine learning. It helps in showing relevant advertisements and promotions to users based on search queries,” said Rastogi. “Next, based on the past sales or festivals, we help sellers with the right promotions, discounts and deals that they should be offering on an event or festival that works best.”
Amazon India, which started operations four years ago, has grown 124% every year, and in the second quarter of 2017, it logged 88% year-on-year growth.
“Using technology, we are also improving our catalogue, which is the overall data quality. At times, product images are quite poor or many attributes like colour or material are missing. In such cases, the platform analyses the product and fills the content which vis-à-vis improves the product search listing,” Rastogi said.
According to Rastogi, using deep learning techniques, there are ways of improving a low-quality image after analysing it. “Similarly, if product titles are not very descriptive, it could enhance the title to render better search results,” he said.
The ecommerce giant is also working on incorporating a queries section on its mobile application. “Sometimes answers are instantly available in the product description, but sometimes it’s complicated, so we are working on how to provide instant answers. We are also looking to make reviewing more conversational, where customers can be suggested better choices,” Rastogi said.
Amazon is also using technology to tackle the problem of fake reviews. “Rationally, one customer will not post 7-8 reviews in a week. A genuine customer will take ample time to review a product or will do it once at a time. There are many such signals that the algorithm looks out for before removing these fake reviews,” Rastogi said.
The company is also looking at providing content in regional languages to expand its reach and enhance engagement in tier II and III cities. “AI can play a big role here when you have to translate lakhs of products’ content from English to regional language,” he said.
Amazon India also plans to register more SMEs online from these smaller towns and cities. The company claims that over 80% of its traffic comes from mobile devices, and in view of this, the company had earlier launched a light app of 2MB for such regions.