The core features of the Synthesys™ data analytics platform are:
- Query Augmentation
- Faceted Navigation
- Contextual Search
- Entity Extraction
- Geotagging
- Entity Resolution
Query Augmentation
Address the largest weakness of Boolean Search without taking away what works
- Recall Improvement: The largest issue with Boolean search is the “circular” problem of needing to know what is in your data in order to retrieve the information that best matches your intent
- Minimal loss of precision: Queries can be tuned and further augmented to contract or expand the domain of selection
- Interpretability of Query Result: Unlike other “black box” search technologies, which rely query augmentation basically works with existing Boolean logic and weighting techniques, the reason for certain results appearing is explicable
Faceted Navigation
Summarize key features in datasets and rapid navigation by leveraging metadata for online query constraint expansion and content summarization.
- Rapid Filtering: Quickly subdivide the pile
- High Level Content Summarization: Display high level valuable entity types to characterize the data.
- Substantially a UI issue: Great faceted navigation UIs presume strong web application development design strengths.
Contextual Search
Segment and retrieve key facts within and across documents.
- Performs context-level search
- Provides ability to search for unspecified terms belonging to a category
- Calculates document relevance based on matched contexts
Entity Extraction
Create Flexible, Transparent, and in a manner Predictable Metadata.
- Create custom categories
- Choose training items from full document context
- Train multiple categories from the same screen
- Generate automatic quality metrics
Geotagging
Automatic inference of objective spatial information from clues in unstructured data. Map-based visualization systems require Geographical Information Systems (GIS) coordinates in order to accurately tag a map with precise location information. Of the vast majority of unstructured documents that refer to locations, only 5% actually contain GIS coordinates. The remaining 95% of documents that mention locations don’t contain proper GIS coordinates. Someone has had to markup each document by hand. Until Now.
In response to the direct needs and requirements of the intelligence community, Digital Reasoning Systems developed a superior location-focused solution based on our patented core intelligence engine.
Entity Resolution
Summarize and Compress Critical Entities for Pattern Discovery.
If all data collected contained global unique identifiers (e.g., a bar coded serial number), then semantic reconciliation would be trivial. But the world collects different features in different ways from the same object. Some systems record me as Jeff Jonas and others Jeffrey Jonas. Sometimes I share a frequent flyer number and no date of birth, and in other places I share a date of birth and passport number. So how many Jeff Jonases are there? Organizations that cannot count unique objects make suboptimal decisions and in the case of the multiple loyalty club accounts, maybe denying a decent customer decent rewards, e.g., had all the points been recognized as one belonging to one account!
It is important to address semantic reconciliation before other analytical processes (e.g., statistical analysis, market segmentation, link analysis, etc.). This is a “first things first” principle because semantic reconciliation makes secondary analytic and computational problems that much easier and that much more accurate.
- Jeff Jonas – head of IBM Entity Analytics and former CEO of SRD (NORA)
