Advancing your NLP with Elasticsearch
A talk at Devfest DC 2019 at 11:45 AM | TRACK #2
When you analyze a large number of documents do you often wish there were a better way to display your results? Elasticsearch + Kibana provide an excellent platform for doing just that. Elasticsearch is an open source, distributed, RESTful search and analytics engine. Did you think it was a tool only for cybersecurity professionals? Think again! As a data scientist, it is my new favorite tool to leverage natural language processing (NLP) and visualization. Join me as I show you how I use Python to take my analyzed data and ingest it into Elasticsearch, and then use Kibana to build an attractive dashboard VERY quickly. Elasticsearch + Kibana will allow you (or any user) to search the data and provide multiple visualization tools; from unique words and time series to geographical plots.
Natural Language processing
I presented a brownbag workshop to the WiDS (Women In Data Science) group at Booz Allen Hamilton on natural language processing. November 18, 2019.
This tutorial will cover Natural Language Processing (NLP) of transcribed TED Talks (speeches). There will be a discussion of the algorithms, followed by exercises hosted in a google collaboratory. Using a bag-of-words model, we will extract topics from the corpus using a variety of methods (Latent Dirichlet Allocation (LDA), Non-negative Matrix Factorization (NMF), Latent Semantic Analysis (LSA)). Visualization of these analyses will be shown using the pyLDAviz library and a dimensionality reduction method, tSNE (t-distributed Stochastic Neighbor Embedding). Advanced-intermediate coding experience is strongly recommended.