In 2009, the Department of Health, part of Taiwan's Executive Yuan, announced the advent of electronic medical records to reduce medical expenses and facilitatethe international exchange of medical recordinformation. An information technology platform for nursing records in medical institutions was then quickly established, which improved nursing information systems and electronic databases. The purpose of the present study was to explore the usability of the data mining techniques to enhance completeness and ensure consistency of nursing records in the database system.First, the study used a Chinese word-segmenting system on common and special terms often used by the nursing staff. We also used text-mining techniques to collect keywords and create a keyword lexicon. We then used an association rule and artificial neural network to measure the correlation and forecasting capability for keywords. Finally, nursing staff members were provided with an on-screen pop-up menu to use when establishing nursing records. Our study found that by using miningtechniques we were able to create a powerful keyword lexicon and establish a forecasting model for nursing diagnoses, ensuring the consistency of nursing terminology and improving the nursing staff's work efficiency and productivity.