Nosto’s Search engine finds search results by matching user’s searched keywords in client’s product data defined in Searchable Fields (Search → Settings). When product data is indexed in Nosto and the Search module is enabled, the Nosto team will set up or help clients set up Searchable Fields and their priorities.
Searchable Fields shouldn’t contain irrelevant data to ensure relevant search results; therefore, if you have long product descriptions, we don’t recommend searching in it. Usually, we suggest searching only in product’s name/title, categories, brand, type, and relevant searchable attributes like color or other essential product parameters, which can sit as custom fields or tags.
Nosto’s search engine constantly processes product data in a backend by optimizing, cleaning, and ensuring search relevance.
Moreover, on a query time, Nosto’s search engine enables advanced AI algorithms to maximize search relevance. Various Natural Language Processing (NLP) algorithms are used to understand the user’s intent. One example would be a stemming algorithm to understand that words with the same stem have the same meaning (e.g., knife - knives, deportivos - de deporte) or a compound words algorithm, e.g., damenjacken - jacken für damen. This makes your search intelligent and automated without a need to teach or configure the search engine’s logic.
Finally, on a query base, the search engine applies optional merchandising and personalization to ensure that search results are relevant and ranked by business needs. These topics are covered in the next chapters.