stormsidali2001/IMRAD-introduction-move-two-sub-moves-classifier

IMRaD Introduction Move 2 Sub-move Classifier

This model is a fine-tuned BERT model that classifies sentences from the "Occupying the Niche" (Move 2) section of scientific research paper introductions into their corresponding sub-moves:

• Outline your purpose(s) and state the nature of your research: Stating the research objectives and approach.
• State your hypothesis or research question you seek to answer: Presenting the main research question or hypothesis to be tested.
• Share your findings: Briefly summarizing the main findings of the research (less common in introductions).
• Elaborate on the value of your research: Highlighting the significance and potential impact of the research.
• Outline the structure that the research paper will follow: Describing the organization of the paper (e.g., sections, chapters).

Parent Classifier:

This model works together with the main IMRaD Introduction Move Classifier: https://huggingface.co/stormsidali2001/IMRAD_introduction_moves_classifier

First, use the parent classifier to identify sentences belonging to "Occupying the Niche" (Move 2). Then, use this sub-move classifier to categorize the specific function each Move 2 sentence serves.

Intended Uses & Limitations

Intended Uses:

• Scientific Writing Assistance: Help researchers and students analyze and improve the structure of their "Occupying the Niche" section by understanding the specific sub-moves they've used.
• Literature Review Analysis: Identify how authors state their objectives, hypotheses, and the value of their research in introductions.
• Educational Tool: Illustrate the sub-moves used in Move 2 to clearly define the research contribution within the niche.

Limitations:

• Domain Specificity: Trained on scientific research papers, so accuracy may be lower on other types of text.
• Sentence-Level Classification: Classifies individual sentences, not the entire Move 2 section as a whole.
• Ambiguity: Some sentences might be challenging to categorize definitively, leading to lower confidence scores.

Training and Evaluation Data

Trained and evaluated on a subset of the "IMRAD Introduction Sentences Moves & Sub-moves Dataset": https://huggingface.co/datasets/stormsidali2001/IMRAD-introduction-sentences-moves-sub-moves-dataset

This model uses sentences specifically labeled as Move 2, further categorized into the five sub-moves.

Training Details:

• Base Model: google/bert-base-cased
• Implementation: TensorFlow/Keras
• Evaluation Metrics: F1 score and accuracy

How to Use

from transformers import pipeline

# Parent classifier
move_classifier = pipeline("text-classification", model="stormsidali2001/IMRAD_introduction_moves_classifier")

# Move 2 sub-move classifier
submove_classifier_2 = pipeline("text-classification", model="stormsidali2001/IMRAD-introduction-move-two-sub-moves-classifier")

sentence = "The findings of this study have significant implications for the field of [your field]."

move_result = move_classifier(sentence)
move = move_result[0]['label']

if move == "Occupying the Niche":
    submove_result = submove_classifier_2(sentence)
    print(submove_result)