New Linguistic Description Approach for Time Series and its Application to Bed Restlessness Monitoring for Eldercare

Area of Science:

• Data Science
• Artificial Intelligence
• Linguistics

Background:

• Time series analysis is crucial for pattern discovery and forecasting.
• Generating natural language summaries of time series data is gaining traction.
• Natural language offers advantages over graphical representations for data interpretation.

Purpose of the Study:

• To develop a novel mechanism for generating linguistic descriptions of time series data.
• To create quality summaries that capture relevant time series features for specific applications.
• To demonstrate the method's customizability across different domains.

Main Methods:

• Utilizing natural language processing and fuzzy logic techniques.
• Developing a mechanism to translate time series features into linguistic descriptions.
• Applying the method to bed restlessness data as a case study.

Main Results:

• Successfully generated linguistic descriptions of time series data.
• Demonstrated the method's ability to capture relevant time series features.
• Showcased the quality and customizability of the generated summaries.

Conclusions:

• The proposed method effectively generates linguistic descriptions of time series data.
• Fuzzy logic and natural language techniques offer a promising approach for data summarization.
• The method is applicable to real-world scenarios, as shown by the bed restlessness data case study.