I’m a Data Science consultant based in London, UK.
I help companies making sense of their data using Data Science and Machine Learning, and bringing their R&D efforts in production.
I’m an international conference speaker and trainer, an author with Packt Pub, and a co-organiser of the PyData London meet-up and conference series.
Backed by a PhD in Information Retrieval, I specialise in search applications and text analytics applications, and I’ve enjoyed working on a broad range of information management and data science projects.
I’m experienced in the design and implementation of applications including enterprise search, semantic search, vertical search, text summarisation, sentiment analysis, text classification/clustering and recommender systems.
Some of the technologies I use on a regular basis include:
– Python
– NumPy, SciPy, NLTK, gensim, pandas, scikit-learn
– Elasticsearch, Lucene, NoSQL, SQL
I’m happy to discuss opportunities for a technical contracting role as well as strategic consultancy.
Follow me on Twitter: @marcobonzanini
Connect via LinkedIn: uk.linkedin.com/in/marcobonzanini
Marco,
I’m doing a side project for my personal website, and it involves language processing and semantic orientation, so I found your blog series and am trying to follow along; basically I adapted your twitter analyzer for reddit, and am using it to analyze my own comments, however I’m not sure I’m implementing it correctly…when it comes to co-occurrence, I’m getting useless combos like ((‘unpaused’, ‘whoever’), 1), etc.
I was wondering if you could help me understand it a little better. I can send over my code and everything if you want to look it over or have the time.
LikeLike