Natural Language Processing

Understanding Human-Machine Interactions

Attivio parses user queries and content to understand intent and meaning by employing capabilities such as parts of speech tagging, lemmatization, synonym expansion, classification, entity extraction, key phrases and sentiment analysis. By understanding user intent and inferring meaning based on a deep understanding of text, Attivio automatically presents the most relevant findings from across information silos so unexpected correlations and insights can be discovered and acted upon.

Natural Language Processing Features

  • Semantic search: uses a variety of signals to understand the user’s intent and handle ambiguity. For example, semantic search understands that when a user searches for “profit,” she would also want to find data sets that reference “net income.” Similarly, if a user searches for “NJ,” they would also return entries for “New Jersey.”
  • Autocomplete: displays relevant and popular search suggestions as users type, saving time and frustration. Attivio bases suggestions on user activity and data analysis, making the autocomplete suggestions highly targeted.
  • Spelling correction: increases the speed and efficiency of search. When an exact match isn’t available, Attivio identifies the closest logical alternative.
  • Lemmatization uses variations on words such as plurals, tenses, genders, hyphenated forms, and more. For example, a search for “running” would return matches for “runs” or “ran.”
  • Faceted search groups items returned from a query into the most relevant subcategories. Users can refine their search by drilling down into a particular group or facet.
  • Advanced syntax increases precision through techniques such as phrase search, fielded search, Boolean matching, and proximity search.
  • Fuzzy matching increases recall and allows for looser matching. Substring and approximate matches allow users to find data sources when they only have partial information or even incorrect information.

Insight in 40+ Languages

In addition to English, Attivio supports over 40+ languages including sophisticated features such as statistical entity extraction and algorithmic entity identification and extraction methods for a wide variety of European, Middle Eastern and Asian languages.


  • Recognition and linguistic capabilities essential for high-quality search
  • Tokenization
  • Stop word removal
  • Lemmatization/stemming
  • Decompounding
  • Recommendations for "more like this" results