I am a computer scientist who specializes in developing large scale internet applications. My main scientific interest is natural language processing. In general, I'm passionate about software design, test driven development, all sorts of modern web-technology and of course Ruby.
I'm curious to exchange and share ideas on how we as developers can leverage science to build smarter applications that provide greater user experience. In this sense, I'd especially love to share my passion for natural language processing with others and discuss what we can gain from making the computer understand natural language.
Natural Language Processing (NLP) is the art and science of making sense of user-generated data. It is a combination of state-of-the-art computer science techniques and linguistics. Being able to analyze plain text data allows us to gain a lot of insights. Popular NLP tasks are text summarization, keyword extraction or automatic extraction of the author's opinion from a text. In the age of social media, only NLP provides solutions to analyze what users are really care about. Companies such as Google or Facebook invest millions in NLP solutions to harvest information from all the data they have been gathering over the years.
In this talk, I will present you a real-world NLP problem. We will discuss this problem from both, the linguistic and the computer science perspective. Throughout the talk, we will develop a processing pipeline to efficiently solve this problem in an automated fashion. An NLP pipeline usually consists of multiple components, each solving one aspect of the problem and presenting its own challenges. Among other things, you will learn how to tackle the following essential NLP problems using JRuby and OpenNLP:
Would loooove to see this. It will take me way back to my university years. Would be cool to see how to do in Ruby the things we did in C
I'd love to see this presentation, as I'm interested in NLP and I haven't seen many resources about this related to Ruby language :)
I will leverage OpenNLP and JRuby as technologies to present a solution to the discussed real-world problem.
Yes please!! :-)
Is this pure Ruby NLP btw or is it using CoreNLP through JRuby wrappers for example?