Case Study-

Developed NLP Pipeline
for a leading Market Research Company

The Client

A leading research company in consumer & retail market offers data, industry expertise, prescriptive analysis, and advisory services. The client is operating in 28 cities, serving 2000 companies, 300,000 stores & across 20 industries in America, Europe & Asia-Pacific. They collect point-of-sale data, track retailers, distributors, and operators to help them measure, predict and improve performance across all channels.

The Business Situation

The client was unable to manage and process the bulk of ambiguous and unstructured data collected from their several surveys. They partnered with Diaspark to develop an automated platform to streamline and process terabytes of data and deploy a semantic engine for text mining and search results.


Diaspark developed a unique NLP pipeline for the computer to understand general text (human language) query and extract the right data for it from the system with intelligence. This platform empowers end users with intelligent decisions based on human-generated text inputs.

Key Functionalities
  • NLP engine operating on semantics to convert English language into SQL statement from the schema integrated within the application
  • Structured application flow –
    – User enters simple English Statement
    – Application processes database model and translated into valid SQL Statement
    – Application executes statement in the database
    – Application will display the result in simple grid format along with the executed SQL statement
    – User Interface to enter data and get results on query executed
    – Back-end solution to process NLP results
Benefits To The Client
  • Intelligent decision making based on text inputs
  • Improved business opportunities identification and addressing concerns through analysis of customer feedback
  • Improved business value with quick answers from large content repositories
  • Increased business outputs with structured results