From Seth Grimes article – NLP: Everyday, Analytical & Unusual Uses
Every business process (or personal need) that involves speech or text with volume, velocity, or complexity sufficient to push you to seek automated assistance can benefit from natural language processing. So let’s review, systematically, what NLP can do for you. What follows are 23 facets, plus examples illustrating implementations and R&D initiatives.
It is heartening to see emerging technologies like Big Data Analytics, Natural Language Processing and Artificial Intelligence combined into powerful tools to solve real problems. IBM Watson is an example. From Wikipedia
Watson is an artificial intelligence computer system capable of answering questions posed in natural language, developed in IBM‘s DeepQA project by a research team led by principal investigator David Ferrucci.
From Watson provides cancer treatment options to doctors in seconds
Watson has ingested more than 600,000 pieces of medical evidence, two million pages of text from 42 medical journals and clinical trials in the area of oncology research. Watson has the power to sift through 1.5 million patient records representing decades of cancer treatment history, such as medical records and patient outcomes, and provide to physicians evidence based treatment options all in a matter of seconds.
Starting with 1,500 lung cancer cases, Memorial Sloan-Kettering clinicians and analysts are training Watson to extract and interpret physician notes, lab results and clinical research, while sharing its profound expertise and experiences in treating hundreds of thousands of patients with cancer.
The core of the technology is available through Apache Foundation’s UIMA Project.
UIMA – a framework for analyzing large volumes of unstructured information and discover relevant knowledge to a user
You can look at sites and services that take advantage of UIMA’s capabilities.
Applog is a series of curated links of innovation applications of emerging technologies.
Coursera has a nice course on Natural Language Processing. I missed it when it started, so catching up now viewing the archives.
What makes Natural Language Processing Difficult?
1. Ambiguity in the language. This slide shows other difficulties.
2. What tools do we need?
My interest in NLP stems from the following:
- A better way to make sense out of a lot of data I get from Tweets. Instead of reading each one (of even a select group) and trying to make sense, I would love to process them a bit and pick some topics and filter them.
- Extract entities from not only the tweets but also the links. Which companies and products are being mentioned? Why?
- Provide a semantic layer on top of some of our information discovery and monitoring tools and increase their value.
You may have your own reasons. Whatever they are, this course may be worth checking out to get an appreciation of problems and some possible solutions.
From Five Myths About Automatic Sentiment Analysis
Sentiment analysis using natural language processing. Yes, it is done by a machine and no, it’s not 100 % accurate. The industry estimates that it’s at 70 – 80%. We are very open about that and recommend that it be used as an overview.
It would take hours to manually review the same amount and one still wouldn’t have an overall sense of the percentage positive vs negative.