This repository is the official implementation of the paper "Semantic Parameter Matching in Web APIs with Transformer-based Question Answering" (see citation) and contains notebooks for fine-tuning and evaluating a pre-trained BERT model to the task of semantic parameter matching in Web APIs. Additional materials (datasets, model checkpoints, and reports of executed notebooks) can be found on Zenodo. The fine-tuned model (best checkpoint with a top-k accuracy of 81.46% for k=1) is available on Hugging Face.

We implemented a Flask application that places our model behind a Web API and UI for inference. To use this application, navigate to tools and create a docker image with:

docker build -t restberta-core .

Start the docker container with:

docker run -d -p 80:80 --name pm-cpu restberta-core

If you want to start the application for another Web API integration task, i.e., with another RESTBERTa model (e.g., for endpoint discovery, see RESTBERTa), specify the model as ENV parameter:

docker run -d -p 80:80 -e MODEL=SebastianKotstein/restberta-qa-endpoint-discovery --name ed-cpu restberta-core

To open the Web UI, use a browser and navigate to http://localhost:80.

Use the following cURL to make a prediction:

curl -L 'http://localhost:80/predict' \
-H 'Accept: application/vnd.skotstein.restberta-core.results.v1+json' \
-H 'Content-Type: application/vnd.skotstein.restberta-core.schemas.v1+json' \
-d ' {
 "schemas":[
    {
      "schemaId": "s00",
      "name": "My schema",
      "value": "state units auth.key location.city location.city_id location.country location.lat location.lon location.postal_code",
      "queries": [
        {
          "queryId": "q0",
          "name": "My query",
          "value": "The ZIP of the city",
          "verboseOutput": true
        }
      ]
    }
  ]
}'

@INPROCEEDINGS{10.1109/SOSE58276.2023.00020,
  author={Kotstein, Sebastian and Decker, Christian},
  booktitle={2023 IEEE International Conference on Service-Oriented System Engineering (SOSE)}, 
  title={Semantic Parameter Matching in Web APIs with Transformer-based Question Answering}, 
  year={2023},
  volume={},
  number={},
  pages={114-123},
  doi={10.1109/SOSE58276.2023.00020}}