news/senseforth-becomes-first-conversational-ai-startup-to-break-into-top-5-of-squad2.0.jpg Becomes the First Conversational AI Startup to Break into Top 5 of SQuAD2.0

Senseforth’s proprietary NLP engine delivers an industry-leading accuracy of 96% recently attained the 5th rank on the SQuAD2.0 leaderboard, beating its previous best of #6. The model “TransNets + SFVerifier + SFEnsembler (ensemble)” submitted by outperformed humans by a significant margin, and ranked above the models submitted by several technology giants. With this achievement, has become the only Conversational AI platform in the world to feature in the top 5 of SQuAD2.0

Senseforth Squad ranking

What is SQuAD2.0?

SQuAD2.0 (Stanford Question Answering Dataset) is a widely recognized, top-level machine reading comprehension challenge in the field of cognitive intelligence. The reading comprehension dataset comprises over 150,000 questions. Unlike other machine reading comprehension tasks, models participating in SQuAD2.0 are not only expected to predict answers to a question in the content of the dataset, but also refrain from answering questions that are not supported by text context.

NLP Engine - the soul of a conversational AI platform

A higher SQuAD2.0 rank is a testament to the superior performance of the NLP engine, which is responsible for interpreting what the user says and converting that language into structured inputs that the system can understand.

AI-powered solutions built by deliver industry-leading accuracy of 96%. This is made possible by the NLP engine that is deeply integrated with A.ware, our proprietary Conversational AI platform. The NLP engine has 3 key modules:

  • Automatic Intent Identification: Also known as ‘intent classification’, it is the task of taking an input and categorizing it based on what the user wants to achieve. Intent recognition is used in conversational commerce, customer service, and other areas.
  • Named Entity Recognition (NER): This allows the NLP engine to easily identify key elements like names of entities, places, people, brands, and more. Extracting the main entities helps sort unstructured data, and detect important information.
  • Sentiment Analysis: Also known as opinion mining, it is the process of determining the emotional tone behind a series of words. It is used to understand the attitude, opinion and emotion expressed by the user.

Can a sub-par NLP engine damage your brand?

A great conversational experience (text or voice) needs an exceptional NLP engine. It has been observed that users fall back on traditional methods of support like phone or email when the chatbots or voice bots fail to address their concerns. That’s not all, when bots misunderstand user queries and provide incorrect information, it could severely dent the brand’s reputation, and add more pressure on the already stretched contact centers.

Conversation gone wrong

The NLP engine built by has outperformed its competitors, and delivers industry-leading accuracy right from day one. In addition, pre-built intent libraries for 10+ industry verticals have further helped improve accuracy across the board.

Learn more about A.ware, our proprietary Conversational AI Platform

Looking to harness conversational AI to automate business processes?