We Use Cookies
This website uses cookies to ensure you get the best experience on our website.
Share on facebook
Share on twitter
Share on linkedin
Enriching metadata for NLP Models
for a SaaS customer engagement solutions
0
M+
Algorithm Queries Solved
0
%
Accuracy Level
About the Customer
Freshworks is a cloud based Customer – Support Software firm that provides SaaS customer engagement solutions. The company’s suite of products is widely used by over 150,000 business firms.
Problem
Freshworks has a Conversation AI interface that receives numerous customer queries every day. Training of the NLP model for Conversational AI requires the intent for the queries to be identified and mapped to specific (FAQs). In addition, Named Entities needed to be specifically identified by the NLP Model. The need was to speed up NLP Models and achieve scalability.
Approach
NextWealth was able to train and validate their machine learning algorithms for all of their customer queries. The process is now happening at a rapid rate with more than 99% accuracy with data services such as Intent identification, Intent mapping, relevancy and ranking of the tickets.
Results
The accuracy delivered by training the NLP Models was beyond customer’s expectation and in turn improving the efficiency of their Conversational AI platform.
Services Provided
Intent Identification and Mapping
Intent Validation of Chats
Named Entity Tagging
Similar Ticket Relevance and Ranking
Let Us Help you with your NLP Requirements
what our Customers say about us
The NextWealth partnership has proven to be of immense value to the Machine learning efforts at Freshworks. What we started off as an experimental program a year ago, has today matured to a level where we involve them ass part of several of our data enrichment workflows. The scope of the partnership includes generating training datasets for ML models, estimating model output quality, enriching metadata for NLP models, and so on. The most impressive thing about NextWealth has been their ability to maintain high levels of accuracy even as their workforce expands and churns over time.
Swaminathan PadmanabhanSenior Director, Datascience @Freshworks 