I'm almost tempted to use #NLTK with #Python or something, and #program a simple #anagram generator myself.

But it's hard to turn my human process for creating anagrams into an #algorithm. There's so much subjectivity in making one, or even recognizing a good one when I see one.

A bit of time on the #NLTK interface this afternoon. Seems like a deficiency in the pos_tag() routine which gives erroneous noun vs proper noun label solely based on capitalization of the first letter. Worked around it okay, but it's a bit disappointing.

Like, I can see it saying "Jack" versus "jack" are an NNP vs NN respectively, but something like "california" should prob always come up as proper noun. An report of that being a verb seems to have gone away at least.

Hmm, to rationalize some of the 'part of speech' weirdness in that bot, I have to look into the natural language tool kit (#NLTK) and WordNet which I haven't touched in a few years.

it's a very fun database, I recall, but really stuffed full of information.

#botSmithing

Ek het 'n konseptuele Afrikaanse stemmer geskep vir gebruik in rekenaar #linguistiek en kunsmatige intelligensie toepassings.
Mens kan die konsep Afrikaanse #stemmer hier uitprobeer:
https://puzzlefoundry.com/TaalStem?lang=af
#NLP #NLTK #Afrikaans

I also keep misspelling #NLTK as NTLK... 😅

Wrote a extremely basic sentiment classifier based on #NLTK for a university project.

Works pretty well on the data set, but I haven't been able to write a sentence myself that didn't result in a "negative" sentiment classification...

Am I that angry? 😂

@dbates Not missing #surveillanceCapitalism of #birdsite, but feats U mentioned’ll find their way into #Mastodon clients or #ActivityPub

Enough 2 collect UR fav. subjects based on UR likes & favs. Then analyze daily posts of people U observe, bin their posts into subjects. Finally display on UR timeline, favorite subjects 1st. #Python #NLTK can do it

Getting people 2 agree is another story. Look @ Dante’s screaming when one of #FediSearch plug-ins started 2 index their timelines

Which #python library do you use for summarizing a bunch of text?

#deeplearning #nltk #nlp #machinelearning

My next few days will be trying to figure out how to use #NLTK and #scikit to make sense of average 3-4GB semi garbage data once a day „as real-time as possible“. Let’s see how this goes :) (I know how to use there tools, but the data is so bad even basic grouping is hard) #python

My #emoji finder was noticeably slow on the first search after restart, which I wanted to speed up.

I was blaming sqlalchemy but it was actually the WordNetLemmatizer all along. The first use takes half a second, so I just do a dummy lemmatization in the init of the class and that greatly improves performance of the first search after restart. #python #nlp #nltk #DataScience

ngrams but for microblogging entries #nltk help

Tonight's work on #Pyzantine [1], a Free tool to analyze ancient greek medical texts.
- A #streamlit widget [2] to make texts clickable (I hate you nodejs)
- Use the experimental memoization for quick interactions when changing words now that the parameters for co-occurrence analysis are dynamic.

It is getting there UI wise, so I'll be able to focus more on the NLP side now.

[1] https://github.com/bjonnh-streamlit/greek_manuscripts
[2] https://github.com/bjonnh-streamlit/st_clickable_text

#python #nlp #pharmacognosy #greek #medicine #docker #nltk

#introduction
Hello!
Freshly minted quantitative PhD looking for work in #DataScience. Shifting from #macOS to #Linux / #PopOS from #System76 (tiling ftw) , which is my daily driver now.
Using #Python, #SkLearn, #PyTorch, #RStats, #nltk, etc. for #ML #DeepLearning, #Statistics, etc.
Interested in the above and #Electronics #Arduino #ComplexityScience #DynamicalSystems #SystemsThinking #Philosophy #Ethics #Economics #Biases in research, etc.
Trying #ComputerMinimalism, i.e. more text-based.

Con #nltk in #python ci vogliono due righe:
https://framagit.org/nemobis/bots/-/blob/master/ngram_tlds.py
https://www.nltk.org/book/ch02.html

Spent a little time today taking another look at wholesome-nltk. It's definitely a good start but the more I talk to people about it, the more things I see to fix 😅 #ai #ml #nlp #nltk

https://github.com/dragfyre/wholesome-nltk

Here's something I like witnessing, socially conscious programming -- Analyzing gender bias in Harry potter books using NLTK and python.

A Bossy Sort of Voice: Uncovering gender bias in Harry Potter

https://youtu.be/JKIbGQn-hoU

#python #nltk