IRL in Media

IBM unveils SCORE/EROCS, helps HDFC Bank Enhance Customer Care

Researchers at IBM India Research Laboratory recently unveiled SCORE (Symbiotic Content Oriented information Retrieval) and EROCS (Entity RecOgnition in the Context of Structured data) technologies which provide an organization the ability to generate insights to help enhance customer satisfaction as well as identify new business opportunities. These two technologies automatically combine structured and unstructured information that are distributed across the organization, enabling organizations to generate new contextual and actionable insights to help determine appropriate business decisions. These technologies are useful for customer information management, targeted marketing, fraud detection and prevention and legal compliance. HDFC Bank has partnered with IBM India Research Lab on this novel first-of-a-kind project to deliver new insight from customer information that will be used to help HDFC Bank deliver superior performance. Information integration will become a key differentiator for enterprises. This new technology seamlessly integrates structured and unstructured data, and exploits context information to optimize efficiencies.

EROCS addresses the problem of linking a document with related structured data in an external relational database. It views the structured data in the relational database as a set of predefined "entities" and identifies the entities from this set that best match the given document. EROCS further finds embedding of the identified entities in the document. These embeddings are essentially linkages that interrelate relevant structured data with segments within the given document. A highlight of this technology, which clearly differentiates it from the traditional named-entity recognition systems, is that EROCS identifies an entity even if it is not explicitly mentioned in the document. It exploits the context information present in the document to match and identify the entities.

SCORE addresses the problem of consolidated querying of structured and unstructured data, wherein the application specifies its information needs using only an SQL query on the structured data, and this query is automatically "translated" into a set of keywords that can be used to retrieve relevant unstructured data. At the core of this technology lies a technique for obtaining these keywords from not only the query result, but also from additional related information in the underlying database. The integration of structured and unstructured data provides HDFC Bank to enrich its business intelligence capabilities, and gain critical insights that can be applied in a variety of scenarios.