Social data provides important insights into the desires, habits and inclinations of customers. While, many platforms survey customer remarks on social media outlets like Facebook and Twitter, such technology is often limited in its “word-spotting” approach which, taken out of context, may yield results irrelevant to a client’s research aims. Converseon’s new text analytics service ConveyAPI combines the speed, processing capacity and scalability of data software with the accuracy, understanding and analyses of social conversations that only a human touch can provide.
ConveyAPI addresses the challenge Dr. Phillip Resnik, a Converseon data scientist, explains: “Where scientific jargon, marketing puffery, and a laundry list of features can often obscure what really matters: using a technology meant to measure human expression, are we obtaining the value of a human analysis?” The program utilizes Converseon’s large database of human coded records of over 50,000 brands in conjunction with “advanced machine learning and natural language processing to attain near-human level analysis at scale, including sentiment, emotion, intensity, relevance, and implicit meaning.” Tests of ConveyAPI’s sentiment coding solution showed the program approximated 90-95% of human performance; the significant percentage suggests the benefits of human coding can be realized cost effectively on a massive scale.
Of course, human language evolves and Converseon has also taken this into account. ConveyAPI is also offering classifiers that hone in on specific business problems, industry verticals, and organizations through continues semi-supervised training of the system. This process involves quickly developing new classifiers and capturing the nuances of human language as it evolves and changes context. Resnick adds that the technology does not need to be perfect, but practical: “So, when building and benchmarking our social media analysis technology, we set our sights on how close our system could get to human performance. One doesn’t need the technology to be 100% perfect, because people aren’t perfect, and we know people can get the job done just fine…The right goal is for the technology to be as good as people.”
In addition to using Convey API for data monitoring, clients can also take advantage of its applications, which range from “predictive analysis for capital markets, lead generation, sales forecasting, eCommerce apps, content personalization engines, and rating and review sites.”
Rob Key, Converseon CEO states: “We see social intelligence as the lifeblood to literally thousands of new and emerging applications. The mission of ConveyAPI is to become the engine that powers an ever-expanding ecosystem of third party socially-intelligent business applications.”
Converseon’s data science team includes Dr. Philip Resnik of the University of Maryland, College Park and Dr. Jason Baldridge, professor at University of Texas, Austin and Co-Founder of the OpenNLP project, who together took nearly three years to develop ConveyAPI.