SCADA-NLI: A Natural Language Query and Control Interface for Distributed Systems

Abstract

Human-Computer natural language interaction is helpful to reduce the operation and maintenance cost of the SCADA system, and it is necessary to solve the complex natural language interface problem that supports data query and real-time control. According to the complexity of natural language instructions, a hierarchical classification of semantic parsing algorithm is adopted. Firstly, the KWECS method is used to classify the intent of natural language instruction, then the TF-IDF keyword extraction algorithm combined with the cosine similarity is used to structure the key-value of the classified natural language instructions which was used into SCADA control intermediate language and then formally converted into actual control or query instruction. If the analysis fails, the complex control and query instruction analysis are carried out according to the classification results, structuring instruction parsing based on dependency parsing and SQL natural language parsing based on deep learning are adopted respectively to implement real-time control interface and database query interface. Our experimental results show that the proposed hierarchical classification of natural language comprehensive query and control interface can better solve the problem of human-computer natural language interaction in the SCADA system, and the accuracy of intent recognition reaches 96.5%. In more detail, the accuracy, precision, recall, and F-score of instruction parsing reach 88.47%, 90.21%, 89.48%, and 89.72% respectively. Especially, it provides more convenient interactive means for industrial and agricultural information management and control.INDEX TERMS Natural language interface, intent classification, semantic parsing, scada systems, humancomputer interaction.

Building a Natural Language Query and Control Interface for IoT Platforms

“…Precision = N ts N s (10) Recall represents the proportion of the number of correct structured logical forms N t s obtained by the corresponding algorithm to the number of all structured logical forms N s d contained in the command, which is calculated as follows.…”

“…This leads to the inability of such models to effectively parse natural language commands with complex logic. The DSA algorithm proposed by Wu et al [10] uses a dependency syntax tree for instruction analysis to extract the '' ACTION''-''LOCATION''-''DEVICE'' triples from the commands to simple manipulation of equipment. We add the rule 3 of semantic parsing algorithm based on dependency parsing to make it applicable to our task.…”

“…Several researchers have explored grammar-based approaches: Noura et al [9] proposed the VISH grammar, which uses multiple rules to extract multiple types of entities from command for analyzing the intent in natural language commands and controlling target devices in smart home scenarios. Wu et al [10] proposed a parsing algorithm based on dependent syntactic analysis, which uses a dependent syntactic tree and defined natural language semantic parsing rules to semantically extract actionlocation-device triplets, making the triads corresponding to each operation logically independent, which solves the problem of parsing commands containing multiple intents to some extent. However, dependent syntactic analysis tools are very sensitive to the length of the command sequence, making it difficult to analyze complex commands with long lengths effectively.…”

A Detailed Study on Aggregation Methods used in Natural Language Interface to Databases (NLIDB)

“…For example, traditional query languages cannot adequately represent the requirements' attributes. Effective user-database interactions [7] are required by many next-generation database systems, which involve intelligent information management. As opposed to working on attribute values, non-expert users increasingly request the ability to query relational databases using more natural language that includes linguistic variables and phrases [8].…”