Finetuned-Distilbert-needmining (uncased)

This model is a finetuned version of the Distilbert base model. It was trained to predict need-containing sentences from amazon product reviews.

Model description

This mode is part of ongoing research, after the publication of the research more information will be added.

Intended uses & limitations

You can use this model to identify sentences that contain customer needs in user-generated content. This can act as a filtering process to remove uninformative content for market research.

How to use

You can use this model directly with a pipeline for text classification:

>>> from transformers import pipeline
>>> classifier = pipeline("text-classification", model="svenstahlmann/finetuned-distilbert-needmining")
>>> classifier("the plasic feels super cheap.")
[{'label': 'contains need', 'score': 0.9397542476654053}]

Limitations and bias

We are not aware of any bias in the training data.

Training data

The training was done on a dataset of 6400 sentences. The sentences were taken from product reviews off amazon and coded if they express customer needs.

Training procedure

For the training, we used Population Based Training (PBT) and optimized for f1 score on a validation set of 1600 sentences.

Preprocessing

The preprocessing follows the Distilbert base model.

Pretraining

The model was trained on a titan RTX for 1 hour.

Evaluation results

Results on the validation set:

F1
76.0

BibTeX entry and citation info

coming soon