Natural language processing (NLP) is a component of artificial intelligence (AI) that enables computer programs and functions to understand human speech as it is spoken. In commerce-oriented websites and apps, NLP supports meaning-based search, allowing shoppers to search for items in their own language while still producing relevant results, even if the search terms do not directly match keywords in product records.
As opposed to pure keyword-based searches, NLP strikes the critical balance between delivering a variety of highly on-target search results without inundating customers with results they’re not looking for.
Here are three tips for successfully applying NLP capabilities to ecommerce:
Focus on Product and Descriptor Awareness
Let’s take the example of a shopper searching for “blue one-piece children’s pajamas.” Basic keyword-based search works by identifying all documents and pages that include one or more of these terms, which could include “blue one-piece children’s bathing suit” or “blue one-piece children’s jumpsuit.” While this may produce a thorough set of results, it actually risks going too far and wide presenting results that may not be relevant, potentially overwhelming and alienating the customer.
Product awareness is the ability to look at a term and identify which word represents the primary item being sought, and which ones are descriptors that have secondary importance. In this case, NLP would be able to see “pajamas” as the primary item being sought. So any results with one or more of the adjectives (blue or one-piece) but are not pajamas would be automatically eliminated from the search results, or ranked further down.
Once the primary product being sought is identified, NLP applies machine learning to understand which descriptors tend to be the most important for individual items. For sneakers, the descriptor may be the exercise type or function; for dresses, it may be style; for t-shirts, it may be color or size. Search results can then be displayed based on the importance rank of various descriptors. If someone is searching for “blue summer dress” and style is identified as the most important descriptor, NLP will recognize which word designates the style, in this case “summer.” A yellow summer dress will therefore be prioritized over a blue cocktail dress in the rankings.
NLP and Linguistic Nuances, Synonyms, Misspellings
NLP in ecommerce site search must be able to recognize similar (though not identical) search terms, based on individual shopper’s unique lexicon preferences and context. For example, one shopper may search for “one-piece blue children’s pajamas,” while another might search for “one-piece blue children’s PJs,” and yet another might search for “blue children’s onesie.” NLP must be able to identify these items as being one and the same thing, producing the same relevant results regardless of the exact terminology being used.
In a similar vein, NLP must be able to identify synonyms, even in cases where the spellings are very dissimilar, i.e. “pants” and “trousers”; “underwear” and “briefs”; “onesies” and “footies.” Also, nothing annoys time-strapped customers more than getting a “no results” page when they search, simply because they misspelled a search term. Most ecommerce sites can’t afford to lose conversions to human errors. NLP can help ensure that a misspelled search for “red jacket” will deliver the same relevant results as a correctly typed search.
Combine NLP with Personalization
Some organizations are combining NLP with greater search personalization, and this is yielding significant benefits. Online customers expect a one-to-one personalized shopping experience, just like they’d have in a physical store. Search can be an excellent vehicle for personalization, prioritizing results based on individual tastes and attributes such as size, color, age, gender, location, brand affinity and style.
The goal here is to implicitly understand a shopper’s preferences from previous site behavior, in-store purchase history and other third-party data sources to create a uniquely tailored experience. Sites that can deliver this while accommodating language nuances or mistakes in search terminology can have a distinct sales advantage.
Applying NLP to search can be incredibly powerful because it switches the focus from keywords to actual meaning, allowing humans to be humans while a machine does the work of accurate, intent-based interpretation. When applied to text-based ecommerce search, the NLP capabilities described above can play a key role in creating the kind of frictionless, seamless shopper interactions that drive ecommerce conversions, in a way that antiquated text-based searches simply can’t hope to.