The project aims to capture humor latent structures such as incongruity, ambiguity, phonetic style and personal affect by building a huggingface transformers model for humor detection.
According to corporate surveys, we have determined that some products have bias in PQA platforms where they attract more humorous questions than others. Naturally, humorous questions are an integral part of PQAs(Product Question Answering platforms) systems. These are web service systems which enable users to post questions and get feedback regarding certain products. Some products attract humor due to their unreasonable price, their peculiar functionality, or in cases that users emphasize their critical point-of-view through humor.
Hypothesis: Classify the customer question on a product as either humorous or Non humorous.
X-Axis: The questions.
y-Axis: The class labels (0, 1)
Experimental setup: Classify the questions on the product as either humoros(1) and non humoros(0) based on the questions customers asked about the product.
Design of the Experiment: Analyze 27916 asked by the customers about avarious products.
Sample size: 27916 questions.
Distributed under the MIT License. See https://github.com/Eltonjohn-Oketch/Humor-Detection-in-PQA-systems-using-ML/blob/main/LICENSE for more information.
Huggingface transformers
Eltonjohn oketch.