GPB and BAC: two novel models towards building an intelligent motor fault maintenance question answering system

Pin Lyu,Jingqi Fu,Chao Liu,Wenbing Yu,Liqiao Xia

DOI: https://doi.org/10.1080/09544828.2024.2335135

2024-04-13

Journal of Engineering Design

Abstract:Generally, the existing methods for constructing a knowledge graph used in a question answering system adopted two different models respectively, one is for identifying entities, and the other is for extracting relationships between entities. However, this method may reduce the quality of knowledge because it is very difficult to keep contextual information consistent with the same entities in the two different models. To address this issue, this paper proposes a model called GPB (GlobalPointer + BiLSTM) which integrates the BiLSTM into GlobalPointer through concatenation operations to simultaneously guarantee the rationality of identified entities and relationships between entities. In addition, to enhance the user experience using an intelligent motor fault maintenance question answering system, a model called BAC (BiLSTM + Attention + CRF) is proposed to identify named entities in user questions, and the BERT-wwm model is used to classify user intentions to improve the quality of answers. Finally, to verify the advantages of the proposed model GPB and BAC, comparative experiments and real application effects of the developed question answering system are demonstrated on our built motor fault maintenance dataset. The experimental results indicate that the constructed knowledge graph and developed question answering system provide engineers with high-quality motor maintenance knowledge services.

engineering, multidisciplinary

What problem does this paper attempt to address?

### Problems Addressed by the Paper

This paper primarily addresses the issues of knowledge graph construction and intelligent question-answering systems in motor fault maintenance. Specifically:

1. **Problems with Existing Methods**: Existing methods for constructing knowledge graphs typically use two different models, one for entity recognition and another for extracting relationships between entities. This separated approach may lead to the loss of contextual information consistency, thereby reducing the quality of the knowledge.

2. **Proposed New Models**:
- **GPB (GlobalPointer + BiLSTM) Model**: This model combines BiLSTM with GlobalPointer to achieve simultaneous recognition of entities and their relationships, ensuring the consistency of contextual information and the rationality of entity relationships.
- **BAC (BiLSTM + Attention + CRF) Model**: This model is used to identify named entities from user questions and uses a pre-trained BERT-wwm model to classify user intentions, thereby improving the quality of answers.

3. **Objective**: By extracting valuable information from large-scale multi-source motor maintenance text data, the goal is to develop an intelligent question-answering system based on the Motor Fault Maintenance Knowledge Graph (MFMKG) to help engineers improve the efficiency of motor maintenance services.

In summary, the main purpose of the paper is to improve the knowledge graph construction methods for motor fault maintenance by proposing new models and to develop a more efficient and intelligent question-answering system.