CCKS 2021 Life Service Domain Knowledge Graph Question Answering Competition
The CCKS 2021 competition invites researchers to develop Chinese Knowledge Base Question Answering systems that leverage a life‑service knowledge graph from Meituan, offering detailed task description, dataset information, registration procedures, timelines, and prize incentives.
Understanding and using natural language is a core AI challenge; knowledge graphs serve as a machine‑understandable language, but their highly structured nature often requires query languages like SPARQL, making knowledge graph question answering (KBQA) difficult for ordinary users, thus KBQA has become a popular research area.
In academia, innovative methods such as semantic parsing and information retrieval flourish, while industry applications like smart speakers, voice assistants, and intelligent diagnosis broaden KG‑QA scenarios, increasing demand for efficient, accurate, user‑friendly, safe, and explainable KBQA systems.
Within Meituan, services such as merchant IM, intelligent customer service, and complex conditional search rely on KG‑QA; together with Peking University, the Meituan Search & NLP team proposed the "Life Service Domain Knowledge Graph QA" task at CCKS 2021, encouraging participants to build innovative KBQA systems that handle both specialized and open‑domain knowledge graphs.
The task, called Chinese Knowledge Base Question Answering (CKBQA), requires a system to read a Chinese question and select one or more entities or attribute values from a provided knowledge base as the answer. Questions are objective fact‑based and may involve sub‑tasks such as entity recognition, relation extraction, and semantic parsing; training can use auxiliary resources, but final answers must come from the given KG.
This year's competition builds on OpenKG, adding a life‑service domain knowledge base (tourism, hotels, food, etc.) sourced from Meituan and merging it with the open‑domain PKUBASE KG, providing both datasets for model training.
Registration is open until July 15, with data released on May 15, test data on July 15, result submission by July 20, and paper submission by August 5. Participants can register via the platform link https://www.biendata.xyz/competition/ccks_2021_ckbqa/ or by email following the specified format.
Prizes include 20,000 CNY for first place, 10,000 CNY for second, 5,000 CNY for third, and an additional 5,000 CNY innovation award.
For any questions, participants are invited to join the competition's WeChat discussion group by scanning the provided QR code.
DataFunTalk
Dedicated to sharing and discussing big data and AI technology applications, aiming to empower a million data scientists. Regularly hosts live tech talks and curates articles on big data, recommendation/search algorithms, advertising algorithms, NLP, intelligent risk control, autonomous driving, and machine learning/deep learning.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.