E-commerce site search engines must be smart enough to recognize product names, categories, product attributes and do it in multiple languages. They also should know synonyms, ignore case, understand syntax, and more.
In this example, from Zalando.co.uk, a user searches for a dark men’s jacket warm and finds winter jackets: dark blue, navy, and so on.
Though the keyword dark is not mentioned in product names, the search returns relevant results. This could be achieved in many different ways, including an AI image recognition algorithm, which can tell the color to configure synonyms in the search engine.
Another example, from ConsigliosKitchenware.com, shows how ecommerce search should effortlessly adapt to both singular and plural forms.
Typos can and are happening especially when it comes to brand names. Even if you do carry the brand, users may not be able to find it if there’s a spelling error. A spell-checker can correct the error to maximize the chances that they’ll get the result they want. Sometimes, the search queries are long and complicated.
Let’s say you have entered “lenovo 16gb ram laptops for business” on Bestbuy.com. The site search is smart enough to return laptops that include: the brand name, the specified amount of memory, which is technically RAM, and labeled “Great for Business,” This demonstrates the search engine’s ability to efficiently juggle multiple specifications from different data fields.
Semantic search is a technique that goes beyond finding keywords. It aims to determine the intent and contextual meaning of the words a person is using for search. This type of search can understand almost precisely what the user is trying to ask.
In this context, semantics refers to the philosophical study of meaning and is strongly linked to Machine Learning. It uses past data and trial-and-error patterns to enhance your user’s experience.
In this example from Bloomreach, search uses natural language processing, advanced attribute extraction, and past visitor behavior to surface the most relevant products. It can distinguish between product and brand name, ingredients, and measurement units.
You can use Semantic search to deliver a truly personalized user experience.