Key Phrase Extraction Ethics Statement Project

What is Key Phrase Extraction ?

• As the name suggests, key phrase extraction is the process of identification and extraction of keywords and key phrases from text. These automatically extracted words/phrases highlight the evident themes/patterns in a given text document. In a sense, they can provide relevant information about the text and help us in understanding the essence of the text. We can broadly classify key phrase extraction as an Information Retrieval(IR) technique.

Why do we need key phrase extraction for the ethics statement project?

• As with any given Information Retrieval task, we aim to find patterns that are characteristic to ethics statements in scholarly articles. Key phrase extraction gives us insights about the practices and policies in medical science etc.

1. We can extract and interpret the types of Consent which a given study adopted for a given medical procedure.
2. We can extract the different ethics committees, organizations, universities and hospitals mentioned in the ethics statements/section of a given publisher.
3. We can access the approval number of the study.

• All of these details assist us in public policy design and implementation. Ethical practices also safeguard the rights of an individual to be informed of the potential risk involved in the procedures/medications that are being administered. The need for such practices is highly evident

PKE models used and link to documentation :-

1. KP-Miner
2. YAKE!
3. TextRank
4. SingleRank
5. PositionRank
6. MultipartiteRank

Other tools and key phrase extraction models:-

1. RAKE Rapid Automatic Keyword Extraction
2. nltk library python
3. String module python
4. Pandas library python
5. Logging module python
6. Spacy library python
7. PyCharm IDE
8. Python version 3.9

Results:

• We obtained the following key-phrases and key words after using the models mentioned above:- https://github.com/petermr/dictionary/blob/main/ethics_statement_project/key_phrase_extraction/results.md

Expectations:

• Example ethics statement :- The studies involving human participants were reviewed and approved by the Institutional Review Board of Tongji Medical College, Huazhong University of Science and Technology. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by the Institutional Review Board of Tongji Medical College, Huazhong University of Science and Technology. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.
• We aim to extract relevant pieces of information such as the organization which is Tongji Medical College in this particular example. This a Named Entity Recognition problem which we have tackled using spacy. We also want to extract the type of consent, in this case it is 'written informed consent'. We can derive information about the involvement of human participants in the study.