Natural language processing-based approach for automatically coding ship sensor data

Abstract

The digital transformation of ship systems requires the coding and management of large amounts of Input/Output (IO) data generated by various pieces of equipment during ship operation. In this study, we investigated a method that recognizes the text of the IO description of a ship to automatically code IO data. Accordingly, the characteristics of the IO descriptions were extracted using Term Frequency-Inverse Document Frequency (TF–IDF) and word embedding, and machine learning techniques such as K-Nearest Neighbors (KNN) and Support Vector Machine (SVM) and deep learning models such as Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and bidirectional LSTM (BiLSTM) were used to classify them into codes. Through the application of different text preprocessing techniques based on the unique characteristics of the data, the performances of the algorithms improved; the experimental results showed an accuracy of up to 91%, with an average improvement in accuracy of 5% for each algorithm.