The Discovery Engine is at the core of our mission to equip data owners with cutting-edge technology to extract the maximum value from their data assets. We recognize that data owners come in many forms and sizes: from startups to global corporations, individual researchers to government institutions. That’s why we architected the Discovery Engine with utmost flexibility in mind.
The Discovery Engine is an in-memory index which specializes in handling large volumes of rapidly changing structured and free text data. We cater to data owners looking for a dynamic, fast product that integrates best-in-class fuzzy matching, structured data navigation and visualization, and free text search.
A Relationship-Oriented Approach to Calculating Relevance
Humans are experts at using the complex relationships among data points to make decisions. Existing search tools, however, tend to treat data in isolation, forcing users to make time-consuming tradeoffs about data attributes themselves. We believe that an intelligent search system should leverage the relationships that humans recognize intuitively.
The Discovery Engine understands how items in a data set relate to one another. These relationships form the inputs to our advanced similarity algorithms that calculate the best matches to a user’s query. Our engine gives data owners a great deal of flexibility in calibrating the weightings on different data attributes, ensuring an optimized relevance algorithm unique to each client.
Read more about how we determine Relevance and Weightings
The Future of Search is “Fuzzy”
Existing search tools typically show only perfect matches to user queries or none at all. These literal search systems produce an ironic result: the more detailed users are about they are looking for, the fewer results get returned. Even in large datasets, it’s not uncommon to receive no results after only a few attributes are selected. These counter-intuitive systems require users to guess at the contents of a data set, wasting time and increasing the probability of abandonment.
The Discovery Engine rewards users for being specific about their requests. Using advanced fuzzy matching, the Discovery Engine always returns the most relevant results, starting with exact matches and moving to close matches. The number of search criteria a user applies will have no affect on the number of results that are returned, eliminating guessing and the frustrating “Zero Results Found” error. The Discovery Engine guarantees that the quality of the results actually increases as users’ queries become more specific.
Read more about how we are innovating with Fuzzy Search
Better User Experience
We recognize that users today have different types of goals when they interact with data sets. Some are “searching” for a particular item they already have in mind. Some have a general sense for what they’re looking for and want help “navigating” around the possibilities. Some are just looking around in hopes of “discovering” something that piques their interest. Some are doing all three.
We aim to deliver an unparalleled user experience across all three types of engagement. In addition to flexible relevance scoring and advanced fuzzy matching, the Transparensee platform also includes the following selected features:
- Best-in-class search speed across millions of records and text entries and hundreds of metadata points.
- Compatibility with interactive front-end technologies like Ajax which enable more engaging and instantaneous user interaction with results.
- Multi-select faceted navigation to guide users to the most relevant data.
- Advanced mapping/GIS and data visualization integration.
- Rich support for tagging and user feedback (e.g. ratings, popularity, etc.).
Read more about our unique commitment to User Experience