Adeptmind, a startup that gives retailers a way to implement a smarter search engine on their sites by collecting related data to all their products and zero in on what customers are actually looking for, has raised $4.5 million from Fidelity Canada to further expand its search tool. G Wo, CEO of Adeptmind, commented that “A lot of times NLP companies will have fairly ‘comprehensive’ knowledge graphs where you do internal labeling, but most of the data comes from the product catalog. As such anything not in the product catalog will not be understood. There’s no free lunch when it comes to machine learning. We target crawl a large portion of the web. Based on the web we do targeted crawling so any relevant information we ingest”.
Yoav Artzi, VP of Products at Adeptmind, was positive about the future and capital raising, and said that “A lot of times NLP services tend to be consulting in nature. You build out a system with people spending three or four months, and then you must do another store and spend another three or four months. Eventually, you’re bounded by linear growth. You don’t have to spend a lot of effort if your system is able to support them through unsupervised learning. We ingest the catalog and get to very high accuracy very quickly. That was harder to do pre-deep learning, so we’re catching the front end of deep learning and NLP”.