Nowadays, population aging has become a prominent problem all over the world, which brings new challenge to manage multi-type and multi-relation health data from elderly. Existing approaches store those complex data in traditional relational database, which lack support of relation retrieval and is hard to answer elderly's semantic query. In this paper, we design and implement a knowledge graph based question-answer platform (KnowHealth) to manage those health data, which can help the elderly know their health condition better, Specifically, we propose an ontology definition which can describes the entities and relations in aged disease domain. Based on the ontology, we crawl and extract health related information from different input sources. Then, entities and relation extraction methods can be used to construct a knowledge graph. Finally, we completed a historical behavior driven question-answering platform to serve query for elderly. By analyzing and extending intention of questions, answer can be retrieved and reasoned more accurately.

The outbreak of Covid-19 has exposed the lack of medical resources, especially the lack of medical personnel. This results in time and space restrictions for medical services, and patients cannot obtain health information anytime and anywhere. Based on the medical knowledge graph, healthcare bot alleviate this burden effectively by providing patients with diagnosis guidance, pre-diagnosis, and post-diagnosis consultation services in the way of human-machine dialogue. However, the medical utterance is more complicated in language structure, and there are complex intention phenomena in semantics. It is a challenge to detect the single intent, multi-intent, and implicit intent of patient’s utterance. To this end, we create a high-quality annotated Chinese Medical query (utterance) dataset, CMedQ (about 16.8k queries in medical domain includes single, multiple and implicit intents). It’s hard to detect intent on such a complex dataset through traditional text classification models. Thus, we propose a novel detect model Conco-ERNIE, using concept co-occurrence patterns to enhance the representation of pre-trained model ERNIE. These patterns are mined using Apriori algorithm and will be embedded via Node2Vec. Their features will be aggregated with semantic features into Conco-ERNIE by using an attention module, which can catch user explicit intents and also predict user implicit intents. Experiments on CMedQ demonstrates that Conco-ERNIE achieves outstanding performance over baseline. Based on Conco-ERNIE, we develop an intelligent healthcare bot, MedicalBot . To provide knowledge support for MedicalBot , we also build a Chinese medical graph, CMedKG (about 45k entities and 283k relationships).