- API docs
- CLI
- Integration guides
- Uploading data
- Downloading data
- Blog
- How machines learn to understand words: a guide to embeddings in NLP
- Prompt-based learning with Transformers
- Efficient Transformers II: knowledge distillation & fine-tuning
- Efficient Transformers I: attention mechanisms
- Deep hierarchical unsupervised intent modelling: getting value without training data
- Fixing annotating bias with Communications Mining
- Active learning: better ML models in less time
- It's all in the numbers - assessing model performance with metrics
- Why model validation is important
- Comparing Communications Mining and Google AutoML for conversational data intelligence
Downloading data
For real-time analytics and automation use-cases we recommend using the , which allows you to iterate through messages in a dataset. If you are integrating Communications Mining as one of the enrichment steps in a data pipeline, take a look at the Predict API routes which may also be suitable for your design.
Datasets can be exported as CSV directly in the browser; there is no size limit, but large files may take a long time to download. We recommend to apply filters before exporting to limit the size of the download and make the CSV file more convenient to work with. Another option for batch download is to use the Communications Mining command-line tool (available for Linux, Mac, and Windows) or the Export API route.
- WHICH DOWNLOAD METHOD SHOULD I USE?
The download methods discussed above will differ slightly with regards to the way they provide predicted labels and general fields. Please be sure to review this comparison table to pick the method that best suits your use-case.
- HOW CAN I USE COMMUNICATIONS MINING LABELS IN AN AUTOMATION USE-CASE?
If you want help to get started with your automation use-case, a step-by-step tutorial using the Stream API is provided here. If you are looking to understand how to use Communications Mining labels in an automation use-case, take a look at the Labels documentation.
- HOW CAN I USE COMMUNICATIONS MINING LABELS IN AN ANALYTICS USE-CASE?
If you are looking to understand how to use Communications Mining labels in an analytics use-case, take a look at the Labels documentation.
