Artificial intelligence (AI) and data mining has been helping Indian ecommerce companies like Flipkart, Myntra, and Lenskart in predicting sales, designing their product lines and developing an interactive platform.
US-based ecommerce leader Amazon is also one of the etailers that is relying on data analytics, AI and machine code to make the buying and selling experience on its platform smooth. After USA, India is the largest technology development centre of Amazon where the data research scientists have been able to create tech solutions for the marketplace.
The machine learning and deep learning experts across Amazon India’s development centre have thorough understanding of computer vision and natural language processing. By performing analytics, the team has been able to address seller concerns and boost customer footprint.
Innovating in India
Amazon India’s CTO Dale Vaz explained how the poor internet connectivity in India compelled the engineers to build Micron, a technology to build lighter version of apps so that consumers from across the country can use it.
Vaz revealed, “We introduced near-automated sign-ups for first-time mobile users, revamped our daily deals offering, removed elements which were too heavy for low end users. The app was reduced from 17MB to 2MB.”
While the less-space consuming app helped buyers, the etailer’s India team also developed a tool for its sellers – Data Mart. This tool has dissected the data into various subsets such as customer preferences, selling history, market trends and sellers’ details. With the help of Data Mart, Amazon not only customizes its targeting strategy but also assist sellers in planning their advertising strategy.
“Based on the data, we help sellers who have lesser experience in online advertising, in putting up ads on select items. Similarly, our systems analyse customer buying data and pitch items to them which they might need in future,” said Vaz.
Artificial Intelligence would also help Amazon to:
- Spot sellers selling overpriced items
- Incorrect product details
- Unearth fake reviews on sites
- Weed out fraud buyers
- Enhance delivery service by looking into location data points
“For instance, a genuine customer who is looking to buy visits Amazon site or app multiple times, views more than 4 products, looks at details of the products and then makes a purchase but certain users just come right in, open one page, book same product multiple times and then choose CoD. In such cases, we block the order to avert frauds,” shared Vaz while explaining how AI can help to spot fake orders.