With the widespread application of artificial intelligence technology, the medical informatization has entered a new stage. Although intelligent question-answering systems have achieved a certain extent of promotion a...
详细信息
Exhibition is an indispensable part of modern society. With the development of social economy and the popularization of Internet technology, exhibition industry has gradually become a new type of industry, which invol...
详细信息
Regional Integrated Energy system (RIES) is an important carrier of Energy Internet. However, due to the complexity of network topology and energy coupling relationship, the operation optimization level of RIES is lim...
详细信息
The proceedings contain 242 papers. The topics discussed include: research on extension innovation model based on the development of smart tourism homestay industry;supply chain management and influencing factors base...
ISBN:
(纸本)9798400709999
The proceedings contain 242 papers. The topics discussed include: research on extension innovation model based on the development of smart tourism homestay industry;supply chain management and influencing factors based on e-commerce delivery data;analysis research and analysis of neural network internet economy under big data modeling – the belt and road initiative as the background;analysis of archive library data system based on genetic algorithm;deep reinforcement learning for boosting individual and aggregate diversity in product recommendation systems;current status and future prospects of brain-inspired intelligence development: a review;and detecting public perceptions of apollo go on Chinese social media based on topic modeling and sentiment analysis.
The article presents an innovative approach to the rehabilitation process for patients with joint endoprosthetics, focusing on the integration of a specialized software module based on computer vision technologies and...
详细信息
作者：
O’Neill, IanSchool of Electronics
Electrical Engineering and Computer Science Queen’s University Belfast Computer Science Building BelfastBT9 5BN United Kingdom
This paper provides insights into the development of a prototype tutorial chatbot, using Azure AI and GPT-4. The prototype system answers and poses questions in the manner of a module lecturer in tutorials – able to ...
详细信息
In the context of the new era, the scale of computer hardware and system complexity began to change, the emergence of super-capacity storage equipment and multi-core CPU in the market, which makes it possible to store...
详细信息
The article deals with development of a control system for the feed drive of metal-cutting machines and machining centers operating under computer numerical control which enhances machining quality by compensating for...
详细信息
WSN system can effectively meet the application of monitoring in terms of CNC status and structural health, but it is difficult to solve the problem of large power consumption. In this paper, we propose an optimizatio...
详细信息
Knowledge graph-based question-answering (KBQA) systems suffer from problems such as low-quality datasets and limited categories of candidate entities, which lead to difficulties in system construction and limited app...
详细信息
ISBN:
(纸本)9798350386783;9798350386776
Knowledge graph-based question-answering (KBQA) systems suffer from problems such as low-quality datasets and limited categories of candidate entities, which lead to difficulties in system construction and limited applications. To address this problem, a low-cost wide-domain KBQA system construction framework called LCWD-QA is proposed by combining deep learning, knowledge graphs, and large language models. First, an entity span prediction model was designed to recognize potential entity mentions in sentences. Then, the concept of entity popularity is introduced, and an entity-linking algorithm is designed to link entity mentions to specific entities in the knowledge graph. Finally, the Bert style of text classification models was used for intent recognition, and then different methods were used to generate the answer to that question. In addition, large language models were used to enhance the experimental dataset and generate supplementary answers to the knowledge graph to improve the generalization of the system. The experimental results show that the LCWD-QA system presented in this paper exhibits good performance on both entity span prediction and intent recognition subtasks, with an accuracy of 99.5%, which is better than that of the prevalent benchmark models. The system avoids the high dependence of traditional named entity recognition schemes on manually labeled datasets and has high accuracy and interpretability, which has high application and reference value.
暂无评论