BERTopic VS sentence-transformers

Try experimenting with different hyperparameters, clustering algorithms and embedding representations. Try https://github.com/MaartenGr/BERTopic/tree/master/bertopic

Try out this notebook which comes with the BERTopic repository.

After some googling, I found this (but for M1 chip Mac) --I wonder if I'm stuck. Is this laptop just not up for the job of working with sentence transformers? Appreciate your advice

Most has already been said and I am not sure how relevant this is but since you are focusing on human raters it might be worthwhile to mention that there is a Pull Request in BERTopic that allows you to use models on top of the default pipeline that further fine-tunes the topic representation. In theory, this would allow you to even use ChatGPT or any of the other OpenAI models to label the topics. From a human annotator perspective, this might be interesting to pursue.

The BERTopic library allows you to plug and play any type of embedding.

A number of excellent topic modeling libraries exist in Python today. BERTopic and Top2Vec are two of the most popular. Both use sentence-transformers to encode data into vectors, UMAP for dimensionality reduction and HDBSCAN to cluster nodes.

BERTopic is a library to consider if you want something that groups data by topic.

I have experimented a bit with BERTopic but didn't find the results very useful. The issue is, that it is very important what exactly people are liking or disliking about the products, not just the fact that they are talking about specific aspects.

txtai is an open-source first system. Given it's own open-source roots, like-minded projects such as sentence-transformers are prioritized during development. But that doesn't mean txtai can't work with Embeddings API services.

Ok great. My use case is not very specific, but rather general. I am looking for a model that can perform asymmetric semantic search for the languages I mentioned earlier (Urdu, Persian, Arabic etc.). I have also looked into the sentence-transformer training documentation. Do you think it would be a good idea to use the XNLI dataset for fine-tuning? Or maybe you can suggest much better dataset. Furthermore, I am not sure if fine-tuning is suitable for my task. Because my use case is general so I can use already trained model.

The task you are looking for is semantic textual similarity. There are a few models and datasets out there that can do this. I'd probably start with the SemEval2017 Task 1 task description and competition entries here and then work outward from there (using something like SemanticScholar or Papers With Code to find newer state of the art works that cite these models if needed). For what it's worth you might find that Sentence Bert (SBERT) gives good vectors for cosine similarity comparison out of the box for this task.

Check out the sentence-transformers implementation. If I don't miss anything they don't exclude CLS when the pooling strategy is set to 'mean'

Break up transcript into shorter segments and convert segments to a 768 vector array. Use a process known as embedding using our second ML model, UKP Labs BERT’s sentence transformer model.

Not sure what kind of texts you have, but these models have a max sequence limit of 512 (approx 350 words or so). If you're texts are longer than that, consider splitting them up into chunks or creating a summary and taking an embedding of that. Some clustering algorithm may be the way to go here. Here's a bunch of examples. I use agglomerative for my use case.

https://github.com/UKPLab/sentence-transformers/tree/master/examples/training/data_augmentation (Augmented Encoding)

For vector representation you can take the mean and then pca to get the size that you want, but if you have time then use sentence transformers to train a vector representation instead.