Description

Digikala online market has recently published some open source data in various categories.

Since I always wanted to do some NLP project, then I thought of some useful tutorials in python for newcomers. I really hope this could be useful for you guys.

I still keep updating the package and also will share the link of video and article related to this post soon!

If you like the content

If you like the content, just add a star. 😏

Before you run models

First you should run the 0 - data Wrangling.ipynb to preprocess the data before going for the rest of files and creating your models.

Requirements

Use these conda commands to install the packages in environment:

conda install -c conda-forge --file requirements.txt

Dataset

I used mini-version of digikala customers comment dataset from here

🔗 www.quera.ir

which was uploaded for a AI competetion on 1398/08/16 and can be found here.

🔗 dataset download.

(Of course Needs authentication 😎).

Full version available in these links:

🔗 source 1

🔗 Source 2

For more studies:

for text preprocessing:

🔗 https://www.kaggle.com/sudalairajkumar/getting-started-with-text-preprocessing 🔗 https://www.kaggle.com/kernels/scriptcontent/19201884/download

tfidf:

🔗 https://towardsdatascience.com/multi-label-text-classification-with-scikit-learn-30714b7819c5 🔗 https://kavita-ganesan.com/tfidftransformer-tfidfvectorizer-usage-differences/#.Xc3OG67ngRY

basic word2vec:

🔗 https://medium.com/explore-artificial-intelligence/word2vec-a-baby-step-in-deep-learning-but-a-giant-leap-towards-natural-language-processing-40fe4e8602ba

gensim:

🔗 https://towardsdatascience.com/machine-learning-word-embedding-sentiment-classification-using-keras-b83c28087456

keras with gensim:

🔗 https://www.depends-on-the-definition.com/guide-to-word-vectors-with-gensim-and-keras/

LSTM:

🔗 https://medium.com/free-code-camp/applied-introduction-to-lstms-for-text-generation-380158b29fb3