Published on: 23.05.2023
Author: Alex Lavrynets
Published on: 23.05.2023
The StatBot.swiss project strives to develop a chatbot able to interact with the Swiss public administration's open data in question and answer format. The project is currently being carried out by the Data Science Competence Center (DSCC) and the Swiss Conference of regional statistical offices (CORSTAT) with the support of the Digital Public Services Switzerland and the participation of the Zurich University for Applied Sciences (ZHAW). A prototype of the chatbot is planned for the end of 2023.
StatBot.swiss: a project based on the Swiss public administration's open data
According to the 2019-2023 Open Government Data strategy (not available in English), the publication of the Swiss public administration's open data "strengthens transparency and participation, promotes accountability and generates added value (efficiency gains in the administration, the encouragement of innovation, and the emergence of new business models)". The central portal opendata.swiss lists the administration's open data sets and makes them available to the public. However, it is still difficult for the uninformed public to explore and use this data effectively. This often requires data handling skills, such as coding, and data formatting. The StatBot.swiss project therefore strives to develop a chatbot able to interact with the Swiss public administration's open data in question and answer format in natural language (language used by human beings). This approach is intended to facilitate the exploration of open digital data. The StatBot.swiss project is supported by Digital Public Services Switzerland.
StatBot.swiss: a collaborative project
The StatBot.swiss project is being carried out jointly by the Data Science Competence Center (DSCC) and the Swiss Conference of regional statistical offices (CORSTAT). The first part of the project was focused on the harmonisation and standardisation of a subset of data from the opendata. swiss website and the aim of this step was to make the data machine-readable. Based on this work, the DSCC - together with the Zurich University for Applied Sciences (ZHAW) - is currently developing a chatbot based on machine learning to efficiently query this database.
StatBot.swiss: an ambitious project
The StatBot.swiss project is based on an innovative open source solution, ValueNet. It was presented by the ZHAW as part of the INODE (Intelligent Open Data Exploration) research project, supported by the European Union. ValueNet offers a complete system for interpreting natural language and translating it into Structured Query Language(SQL). It then allows the search to be launched following the query's specific terms in a relational structured database (data organised according to their logical relationships). ValueNet allows the user to query the database without any knowledge of programming and schema (the logic behind the relational organisation of data). The chatbot will be programmed to handle requests in German and English initially with French and Italian to be supported in the future. A prototype chatbot, in the form of a minimum viable product (MVP), is planned for the end of 2023. The program's source code will then be made available to the public.
Federal Statistical Office
Data Science Competence Center DSCC
Espace de l'Europe 10
Comments on the blog
For feedback on our blog, please use the form below.
Thank you very much!