Key Terms: AI for Intelligent Search

There’s a tremendous amount of buzz these days around Artificial Intelligence, and the concepts and techniques associated with it. These concepts and techniques involve sophisticated technology, and their explanations are often confusing to a non-technical audience. But we'd like to help you better understand what an AI search engine is.  

Dictionary

Key Terms Related to an AI Search Engine

To make the explanations more accessible to the layperson, we’ve created a list of definitions for a number of key AI terms related to an AI search engine.

Artificial Intelligence (AI): AI is an area of computer science focused on enabling machines to act and react as if they were human. We encounter aspects of AI in day-to-day life when we search for something on Google, interact with a voice response system, or download an e-book based on an Amazon recommendation.

Enterprise Search: Enterprise search is an approach that provides an organization the ability to help people (generally the organization’s workforce, but often customers as well) access information contained in databases, document management systems, email systems, messaging systems, and other repositories within an organization’s walls.

Cognitive Search: Cognitive search represents “next generation” enterprise search. Rather than relying on simple keyword search, cognitive search uses AI technologies to gain a better understanding of a user’s intent in making a query, and to provide a more relevant search result.

Machine Learning (ML): ML is what makes the “intelligence” in AI more intelligent. ML programs keep track of what people search for, what they click on, and whether they get the information they need. As data mounts up, a ML program is able to reverse engineer an algorithm based on the accumulating data.

Natural Language Processing (NLP): NLP is an AI technique that translates human communication so that it’s understandable to a computer in the same way that it would be to another human. NLP parses a search request so that it deeply understands the user’s intent and meaning. With this level of understanding, search can respond with the most relevant findings.

Text Analysis: If NLP handles the front end of search, taking in the human conversation and translating it so that the computer understands it, text analytics classifies, groups, clusters, and mines that information for concepts and patterns. Using a variety of techniques, text analytics categorizes the underlying content and is able to come up with the relevant answers.

Sentiment Analysis: Sentiment analysis identifies how a customer feels about a company’s products, services, or brand. Using statistical classification, it evaluates feedback to determine whether a statement is positive, negative, or neutral, and to determine how these statements relate to terms or concepts.

Search Analytics: Search analytics applies statistics to search data itself, identifying trends (most frequent searches, most valued results). Search analytics is used to help understand and optimize performance.

Entity Extraction: Entity extraction is the task through which textual entities (elements) are detected and classified into pre-defined categories, such as people, companies, locations.

Query-Time JOIN: Attivio’s patented Query-time JOIN is a technique that enables users to access and analyze all types of information, regardless of data source or type. It enables you to unify all your structured and unstructured data across every enterprise application – without flattening any data – thereby maintaining your existing data relationships. And it can all be done without building a data model (schema) in advance.

Ready to discuss in more detail?  Contact us - we'll be happy to talk through any of these terms or discuss your current needs.  

Talk to Attivio
Ready to talk to us about how intelligent answers and insights can help your business? Book time to talk with an Attivio expert.