The aim of this repository is to centralize the work performed by the authors to write the paper: "From Static to Intelligent: Evolving SaaS Pricing with LLMs" for SOC4AI Workshop for ICSOC 2024 about an AI-driven solution to model pricings with Pricing2Yaml syntax.
The repository contains one folders:
output
In addition, there is a requirements.txt file that contains the dependencies needed to run the scripts, a LICENSE file that contains the license of the repository, a README.md file that contains the information of the repository, and four .py files: main.py, WebDriver.py, GeminiAPI.py and Yaml2Class.py. Moreover, there is an excel, results.xlsx, that contains the results of the validation of the original paper. Lastly, there is a .pdf file, ICSOC24_Workshop_SOC4AI__Technical_Report.pdf, that contains the technical report of the original paper.
This folder contains the outputs generated by the script when executing it for the following 30 SaaS:
| SaaS |
|---|
| ClickUp |
| Microsoft 365 |
| Salesforce |
| Slack |
| Buffer |
| Hypercontext |
| Notion |
| OpenPhone |
| Pumble |
| Tableau |
| Quip |
| MailChimp |
| Deskera |
| UserGuiding |
| Crowdcast |
| Planable |
| Databox |
| Trustmary |
| Evernote |
| Canva |
| Clockify |
| GitHub |
| Figma |
| Jira |
| Dropbox |
| Overleaf |
| Postman |
| Zapier |
| Box |
| Wrike |
Dependencies can be installed using the following command:
pip install -r requirements.txtThe scripts use the following environment variables:
GOOGLE_AI_STUDIO: The API key for the Gemini API.
You should create a .env file in the root of the repository with the following content:
GOOGLE_AI_STUDIO = YOUR_API_KEY
You can get your API key by following the instructions in Gemini API.
Warning
The scripts is designed to be run with Python 3.9 or higher. Python 3.12.4 was used to develop the scripts. For using the Gemini API, you must take into account rates and limits as well as the region where you are going to use it.
The scripts can be run by running the file main.py. The file contains two constants that should be modified to run the scripts:
URL: The URL of the SaaS pricing.SAAS_NAME: The name of the SaaS.
Reading the original paper and/or the technical report is recommended to understand the limitations of the scripts.
This work is licensed under a Creative Commons Attribution 4.0 International License.
