butterm

Thanks for doing this. I wanted to do the same things a few months back. I looked into a lot of dictionary APIs, but as you mention in the repo, they suck at connecting different parts of speeches. It's funny how simple this sounds but how difficult it is to actually do it. Back then, I gave up and went with a Lemmatizer. Will definitely use this.

I am using elementary OS for the past one and a half years, and honestly speaking, I am in love with it. I am a long time Linux user, and whenever I get a chance, I tell people about all the great things about Linux and why it's better than Windows and the MacOS. But there was always one aspect of Linux that I always felt a bit uneasy about, and this was design/UI/UX/accessibility of the many popular distributions. I liked ubuntu the most in this regards, but after they moved to Unity, I had to ditch it. But then I found Elementary, and have never looked back. Elementary combines the freedom and transparency of open source with beautiful design. I have never seen an OS that scores high on so many different factors.

There's a lot of discussion here about the stock apps in elementary. However, since I am an experienced Linux user, I don't really care about stock apps. The Linux app ecosystem is extremely diverse, and over the years, I have found my best solution for each task. For example, Clementine for music, Atom for text editing etc. So these apps get installed immediately after every installation of elementary and then I never look back at the stock apps again.

Interesting article but that PDF is 39 MB and took an awful lot of time to load on my browser.

Key changes:

* Faster indexing

* Ingest node

* New scripting language (called Painless)

* New data structures from Lucene

* Instant Aggregation

* TF/IDF -> BM25 to calculate relevance

* Fail faster philosophy

* a low level Java HTTP/REST client

I am so glad that you guys open sourced displaCy. I would love to give it a spin on my system. Kudos for all the great work you are doing!

Yup, I know about that but Displacy is just so much more beautiful.

Also, while NLTK's basic Twitter tokenizer is okay, I find that ARK's tokenizer [0] is much better. Similarly, for POS tagging of tweets, I am using the GATE POS tagger [1]. They have a Stanford model and I can hook it up with NLTK using the StanfordTagger class. In fact, this is the kind of integration that I am missing in Spacy.

[0] https://github.com/myleott/ark-twokenize-py [1] https://gate.ac.uk/wiki/twitter-postagger.html

What i love about spacy is their dependency parsing visualization tool[0]. Its so much better than what Stanford offers.

Other than that, I find Spacy's philosophy of "one (best) way of doing everything" a bit stifling. I don't think there is a "best" parser or "best" named entity recognizer. A certain parser may perform very well in a domain (for example, Tweeboparser [1] performs well with tweets) and perform very badly in another. This is true for almost everything in NLP, and NLTK embraces this diversity quite well. This is why NLTK is my go to tool when I want to do something cutting edge in NLP.

[0] https://demos.explosion.ai/displacy/ [1] https://github.com/ikekonglp/TweeboParser