The Salesforce Research team is trying to catch the essence of natural language processing with the help of modern devices. The research team have come up with best kind of device to measure the nuances of natural language processing. In a recent newspaper article, the team has described there intention on research of Natural Language Decathlon (decaNLP).
Richard Socher the chief scientist at Salesforce, he is also leading the team on this research said in his statement “natural language processing is going to open doors for machines whereas understanding language is becoming more difficult day by day because it not only needs a person to understand that particular language but also requires intelligence as visual recreation, to unveil the emotions associated with that language and also the logical areas”.
“We have come across such confusion in interpretation and increase in complexity with choosing the correct word, identifying the grammar, in which context the sentences are being made, tone is very essential to understand the language, sentiment of the context is also necessary, humour, culture reference through the language and there are many more things to keep in mind. Humans will take years to excel so now you can imagine the difficulty of teaching a computer system to know and understand various other languages in just a single unified model. I have given my career to understand and solve this issue. ” Richard added this.
Deep learning has helped many natural language processing (NLP) to emerge as individual. On the other hand general NLP models are not helping to focus on single metric, dataset and tasks. Now we have Natural Language Decathlon (decaNLP) to do the work:-
1)common sense pronoun resolutions
3)semantic role labeling
4)natural language inference
“DecaNLP has the power to change the understanding of NLP community and focusing on alone tasks. Similarly like ImageNet, there are a lot of new research going on, and this will give us a chance to think more broadly about new types of architecture “. Richard said.
“I believe after giving a powerful single default NLP model will inspire other programmers and scientists without doing deep NLP expertise and make progress fast with new NLP and take languages and challenges. This will lead to understand the collaborative language across the globe for business and tools. ”