Voyager Search's Addition of Natural Language Processing Creates a Smarter Search Experience

Jun 15, 2017

One of the key new features added to Voyager Search™ version 1.9.9 is Natural Language Processing (NLP). As a subset of Artificial Intelligence (AI), NLP is the science of teaching computers to understand human interaction and the nuances of language, providing platforms like Netflix and Amazon with the ability to recognize the difference between negative and positive reviews. For our customers, this presents a huge advantage since they will now be able to be even more targeted in their searches using metadata.

With NLP, we’ve been able to make our search engine even smarter. This means better, faster results and a more productive, more efficient experience for our users. Voyager™ software now recognizes the nuances of a location - city, state, company name, company name at another location, etc. - which helps with the search process significantly as it analyzes the words with their meta information. The new capabilities work with the semantics behind the words in order to provide more options for the users to search for what they want.

NLP helps reduce false positives in Voyager’s search results, as well as in our geotagging service, allowing Voyager software to tell the difference between “British Petroleum” and “petroleum”. The NLP service can also be run as a pipeline step on any text field you are indexing and will identify the following:

  • People’s names
  • Nationalities, religions, political groups
  • Facilities, such as buildings, airports, highways, bridges, etc.
  • Companies, agencies, institutions, etc.
  • Place names, such as countries, cities, states
  • Geographic locations, such as mountain ranges, bodies of water
  • Products, such as vehicles, weapons, foods, etc.
  • Events like hurricanes, battles, wars, sports events
  • Art works like titles of books, songs,
  • Laws
  • Languages

According to Dr. Rutu Mulkar-Mehta, CEO of Ticary Solutions, a developer of customized NLP solutions, “Any company that has a lot of data can benefit from automating the process of understanding that data. Let’s face it, humans are limited on both time and availability to look at data. Machine assisted tools tell you the part of the text you should be interested in, enabling you to then focus on more compelling tasks.”

Chatbots, which are computer programs designed to simulate conversation with human users via phone over the Internet, are a great example.

“This is not to replace human interaction,” Dr. Mulkar-Mehta continued. “But it can be very useful in assisting with that. For example, if someone called an automated line really upset, the chatbot could understand they weren’t doing well and mark them as urgent so they could  move to top of the queue. These nuances get lost in some of the current systems, which lead to a myriad of inefficiencies.”

Our customers have been asking for a way to make their search experience even smarter. We’re excited to be able to bring this functionality to fruition and are looking forward to hearing about the unique ways in which they’re using it.

Web Design & Web Development by LVSYS