“Machine learning is ubiquitous on Amazon today,” said Rajeev Rastogi, Vice President of Machine Learning at Amazon India, in an interview with Gadgets 360. demand for products, and improve the quality of product catalogues, categorize products, and also eliminate duplicate products. ,
One of the most basic examples of how heroine is using machine learning (ML) is when you misspelled a query on its search bar. The e-commerce site, Rastogi noted, looks at the phonetic distance between a typed misspelled query and the correct query, rather than looking at their textual distance to provide accurate results – even if you misspelled something.
For example, if you type “geyser” on Amazon to see the available geyser options, Marketplace will autocorrect the spelling and show you the relevant results. Amazon is also using the ML model to translate the content on its site Indian languages that now support it,
Of course, these kinds of computer uses have become common now, and it’s not something most of us think of when we consider the phrases artificial intelligence (AI), or machine learning.
Rastogi revealed that his team is currently working on a seed initiative that aims to bring an interactive shopping experience. It is aimed at first time online shoppers who are more familiar with communicating with offline shoppers when ordering through an e-commerce site.
Conversation commerce, through chatbots, smart assistants like Amazon’s own Alexa, is one of those ideas that keeps coming back every few years as technology improves, and Rastogi talks about that. How it would start with text in English, but evolve into other languages, and to voice.
“A machine can read a document and then answer any question about the document, it’s difficult. For example, today AI can’t review a movie… It is also a challenging problem to summarize a set of solutions. It has not been solved by AI in any way,” Rastogi underlined.
aye Has been used to analyze text and speech at different levels. But computer engineers and data scientists have yet to find a relevant mix to use AI and machine learning to generate accurate assessments, such as movie or product reviews. This is how AI can be used in the literature review process, in a research article published by researchers Gerrit Wagner, Roman Lukyanenko and Guy Pare from HEC Montreal’s Department of Information Technology. famous Even “technically correct tools (such as researchers)” sometimes struggle to evaluate information from sources that use vague, confusing language and presentation.
McKinsey Global Institute (MGI) also partnered with Michael Chui, James Manika and Mehdi Miramadi Told One article states that AI models “have difficulty taking their experiences from one set of situations to another” and require companies to train the models even though the use cases are very similar. This adds additional resource requirements.
Shreyas Sekar, assistant professor of operations management at the University of Toronto Scarborough and Rotman School of Management’s Department of Management, said the effectiveness of AI-based bots in communicating with humans and giving them appropriate results, especially in markets including India. , Sekar did it extensive research How e-commerce platforms are using machine learning, both consumer and warehouse, to scale their operations.
“When you ask these chatbots, simple questions like, No, is it going to rain tomorrow? Or can you play me the song of this movie? They’ve done a great job. But as you get more and more complicated Questions start pouring in, like hey, can you help me find a good shoe for my trek? I guess it’s too hard for chatbots or even . Alexa What is your intention to break this question down explicitly? What are you as a person and how are you different from other people? And which products are matched for you?” he said.
Dealing with Prejudices and Errors
One of the biggest challenges of using AI and ML nowadays is limiting bias and errors. company Google And Facebook To Microsoft Dealing with these mistakes regularly. Amazon is too not foolproof on that Front,
Secker from the University of Toronto Scarborough and Rotman School of Management noted that Amazon’s AI deployment includes a lot of biases that the company is already aware of and is apparently working toward solving them, but it’s not a good idea to work with them. It is not clear how successfully it has achieved the desired results.
“For example, perhaps historically, users have clicked on a particular brand of earphones, so what happens is that in the future, I keep on expanding that exact brand over and over again. So, it’s usually called some sort of earphone. There’s called popularity bias where I try to spotlight products that are already popular, and I’m basically in the system helping the wealthy get rich,” he noted.
However, Rastogi strongly disagreed, and said that Amazon’s goal is to assist human workers, not replace them entirely.
Whom does it help?
The use of AI and ML helps Amazon understand your purchase behavior and purchase history. It is, however, sometimes leads to impulse purchases And just convinces you to buy something you don’t really need. Experts believe that this will further increase with a more interactive shopping experience.
“I think AI and ML can certainly enhance the idea of turning window shoppers into regular shoppers,” Sekar said. “And it’s definitely something that I think is a good way to think of Amazon as a very persuasive seller.”
Consumers themselves can overcome this behavior by understanding how algorithms can influence their choices.
“Even though we’re the ones who click on a product to make a purchase in the end, we’re guided by algorithms in different places along the shopping funnel, whether it’s a recommendation or a review,” Seker said.
Ankur Bisen, senior partner and head of consumer, food and retail divisions at management consulting firm Technopak, said how Amazon uses its algorithms to entice consumers to buy more was exactly the same as in advertising, marketing and marketing. Even discounts. Retail shop done.
“Amazon is doing it with great precision as it’s defined,” he said. “Conversational AI is not the only domain possessed by Amazon’s monopoly. Yes, they are very good at it because of Alexa. But you will see conversational AI emerge in a variety of forms offered by other tech platforms.”