QRMine

/ˈkärmīn/

QRMine is a suite of qualitative research (QR) data mining tools in Python using Natural Language Processing (NLP) and Machine Learning (ML). QRMine is work in progress. Read More..

What it does

NLP

• Lists common categories for open coding.
• Create a coding dictionary with categories, properties and dimensions.
• Topic modelling.
• Arrange docs according to topics.
• Compare two documents/interviews.
• Select documents/interviews by sentiment, category or title for further analysis.
• Sentiment analysis

ML

• Accuracy of a neural network model trained using the data
• Confusion matrix from an support vector machine classifier
• K nearest neighbours of a given record
• K-Means clustering
• Principal Component Analysis (PCA)
• Association rules

How to install

pip install qrmine
python -m spacy download en_core_web_sm

Troubleshoot

• There is a (temporary) issue because of Tensorflows dependency on numpy fixed to < 1.20. Ignore the warnings about incompatibility with TF.
• If there is still an error, install numpy separately as below.

pip install --no-deps numpy==1.20.1

Mac users

• Mac users, please install libomp for XGBoost

brew install libomp

How to Use

• input files are transcripts as txt files and a single csv file with numeric data. The output txt file can be specified.

• The coding dictionary, topics and topic assignments can be created from the entire corpus (all documents) using the respective command line options.

• Categories (concepts), summary and sentiment can be viewed for entire corpus or specific titles (documents) specified using the --titles switch. Sentence level sentiment output is possible with the --sentence flag.

• You can filter documents based on sentiment, titles or categories and do further analysis, using --filters or -f

• Many of the ML functions like neural network takes a second argument (-n) . In nnet -n signifies the number of epochs, number of clusters in kmeans, number of factors in pca, and number of neighbours in KNN. KNN also takes the --rec or -r argument to specify the record.

• Variables from csv can be selected using --titles (defaults to all). The first variable will be ignored (index) and the last will be the DV (dependant variable).

Command-line options

qrmine --help

Command	Alternate	Description
--inp	-i	Input file in the text format with Topic
--out	-o	Output file name
--csv		csv file name
--num	-n	N (clusters/epochs etc depending on context)
--rec	-r	Record (based on context)
--titles	-t	Document(s) title(s) to analyze/compare
--codedict		Generate coding dictionary
--topics		Generate topic model
--assign		Assign documents to topics
--cat		List categories of entire corpus or individual docs
--summary		Generate summary for entire corpus or individual docs
--sentiment		Generate sentiment score for entire corpus or individual docs
--nlp		Generate all NLP reports
--sentence		Generate sentence level scores when applicable
--nnet		Display accuracy of a neural network model -n epochs(3)
--svm		Display confusion matrix from an svm classifier
--knn		Display nearest neighbours -n neighbours (3)
--kmeans		Display KMeans clusters -n clusters (3)
--cart		Display Association Rules
--pca		Display PCA -n factors (3)

Use it in your code

from qrmine import Content
from qrmine import Network
from qrmine import Qrmine
from qrmine import ReadData
from qrmine import Sentiment
from qrmine import MLQRMine

• More instructions and a jupyter notebook available here.

Input file format

NLP

Individual documents or interview transcripts in a single text file separated by Topic. Example below

Transcript of the first interview with John.
Any number of lines
<break>First_Interview_John</break>

Text of the second interview with Jane.
More text.
<break>Second_Interview_Jane</break>

....

Multiple files are suported, each having only one break tag at the bottom with the topic. (The tag may be renamed in the future)

ML

A single csv file with the following generic structure.

• Column 1 with identifier. If it is related to a text document as above, include the title.
• Last column has the dependent variable (DV). (NLP algorithms like the topic asignments may provide the DV)
• All independent variables (numerical) in between.

index, obesity, bmi, exercise, income, bp, fbs, has_diabetes
1, 0, 29, 1, 12, 120, 89, 1
2, 1, 32, 0, 9, 140, 92, 0
......

Author

• Bell Eapen (McMaster U) | Contact |

• This software is developed and tested using Compute Canada resources.

• See also: 🔥 The FHIRForm framework for managing healthcare eForms

• See also: 👀 Drishti | An mHealth sense-plan-act framework!

Citation

Please cite QRMine in your publications if it helped your research. Here is an example BibTeX entry (Read paper on arXiv):

@article{eapenbr2019qrmine,
  title={QRMine: A python package for triangulation in Grounded Theory},
  author={Eapen, Bell Raj and Archer, Norm and Sartpi, Kamran},
  journal={arXiv preprint arXiv:2003.13519 },
  year={2020}
}

QRMine is inspired by this work and the associated paper.

Demo