This site is not optimized for Internet Explorer 9 and lower. Please choose another browser or upgrade your existing browser in order get the best experience of this website.

Understanding the Basics of GraphGrid Search

Understand the basics of GraphGrid Search and how it combines connected data, NLP and ElasticsearchGraph-based search isn’t new – after all, most major internet search engines already use graph technology – but it hasn’t always been readily available.

Twenty years ago, tech giants like Google and Facebook had to build their own graph databases from scratch. Today, developers can use graph search capabilities right out of the box – and enterprise organizations are catching up fast. This next generation of enterprise search engines runs on connected data so that queries are more efficient and search results are more relevant than ever before.

Of course, the easiest way to harness the power of connected data is to use GraphGrid. Here’s how GraphGrid Search helps you build your own powerful search capabilities – without re-inventing the wheel.

Provide a Smoother Search Experience

Before diving into how to build a better search engine, it’s worth noting why the time has come for graph-based search tools.

When your users, data analysts, or other personnel use your search engine, you want them to have a smooth search experience. You need search capabilities that do more than just match keywords in an index – which frustrates users when keywords aren’t an exact fit. Rather, your search tool should be able to quickly find relevant answers to user queries on the first try – and perhaps include intelligent recommendations to other resources. This level of search experience is only possible when you leverage connected data.

With GraphGrid, you build an augmented search that makes broad connections across your disparate data sources while simultaneously narrowing results around the essential meaning of the given search terms. (For example, when a user query includes the abbreviation “St” the search engine should be able to parse from context if the user means “street” or “saint”.) As a result, your users can employ natural-language search terms instead of fiddling with variations on a single keyword over multiple searches.

This smooth search experience is powered by a graph database (like ONgDB) paired with GraphGrid’s Natural Language Processing (NLP) capabilities. When your search engine can understand the meaning of the text and connect that meaning across data sources and silos, your end user finds exactly what they’re looking for without frustration. Furthermore, an intelligent recommendation engine – which is also powered by graph technology – on your search results page provides the user with additional paths for data discovery.

Building Your Graph-Powered Search Engine

With GraphGrid Search you can manage and fine-tune every parameter of your graph-powered search engine.

GraphGrid Search integrates ONgDB with Elasticsearch. This allows Search to send and store connected data into Elasticsearch. This also means that your graph data has all the searching advantages of Elasticsearch.

With GraphGrid Search, you can define policies for populating indexes based on customizable Elasticsearch documents that also support Geequel™. These policies are flexible to your organization’s evolving needs and mission.

In addition, the Continuous Indexing pipeline is capable of event-driven indexing for nodes that are added to your graph over time. This allows you to keep your search engine up to date and sync results with changes as they occur.

Other notable benefits of GraphGrid Search include:

  • Easy: Plug-and-play search index policies and endpoints make it quick and easy for you to set up and start experimenting.
  • Customizable: A highly customizable indexing policy gives you complete control over how indexes are created and structured with Elasticsearch.
  • Analyst-Friendly: Complex search queries can be saved and reused across your organization.
  • Versatile: Diverse options for searching include custom fuzziness, hard filters, sorting, and text suggestions.
  • Fast: Search pre-processes everything that returns. Processing is done ahead of time so results return almost instantly.
    Contextual: Use the graph to narrow, expand, filter, or enhance the initial natural-language results of your search.

To take a deeper dive into GraphGrid Search and other capabilities of GraphGrid, check out the Search documentation or the GraphGrid Search tutorial for beginners.

Conclusion

With GraphGrid, the opportunities for search, recommendations, and data discovery are endless. Using the power of graph database technology plus Natural Language Processing and Elasticsearch, you can build an augmented search engine that meets even the most demanding criteria.

More importantly, when your end-users can find exactly what they’re looking for – or encounter recommendations that exceed their original query – your organization is one step closer to achieving its mission. That’s the power of graph-based search.

Get started today:
Download GraphGrid and try out Search for yourself.