KR-20260067892-A - TIME SERIES DATA PROCESSING METHOD FOR ARTIFICIAL INTELLIGENCE AND SYSTEM SUPPORTING THE SAME

KR20260067892AKR 20260067892 AKR20260067892 AKR 20260067892AKR-20260067892-A

Abstract

A time series data processing system for artificial intelligence according to one embodiment of the present invention includes a data input unit that receives time series data, a data splitting unit that splits the received time series data, a summary data generation unit that generates summary data based on the split time series data, and a data prediction unit that generates a lookback window based on the generated summary data and inputs it into a prediction artificial intelligence model to predict subsequent time series data.

Inventors

정성원
이혜성
김윤영

Assignees

서강대학교산학협력단

Dates

Publication Date: 20260513
Application Date: 20241106

Claims (12)

Data input section for receiving time series data; A data splitting unit that splits the above-mentioned time series data received as input; A summary data generation unit that generates summary data based on the divided time series data; and A time series data processing system for artificial intelligence comprising a data prediction unit that generates a lookback window based on the generated summary data and inputs it into a prediction artificial intelligence model to predict subsequent time series data.
In claim 1, The above data splitting unit is a time series data processing system for artificial intelligence that splits the input time series data at predetermined time granularities.
In claim 2, The above summary data generation unit is a time series data processing system for artificial intelligence that generates the summary data by statistically processing time series data divided for each time granule.
In claim 1, A time series data processing system for artificial intelligence, wherein the above statistical processing performs at least one of the mean, median, minimum, and maximum values for time series data divided for each time granule.
In claim 1, The above data prediction unit is a time series data processing system for artificial intelligence that predicts subsequent time series data using at least one of a neural network, a decision tree, and a support vector machine.
In claim 1, A time series data processing system for artificial intelligence, wherein the data prediction unit generates a lookback window based on the generated summary data and time series data for the summary data, inputs it into a prediction artificial intelligence model, and predicts subsequent time series data.
Step of receiving time series data; A step of dividing the above-mentioned time series data received as input; A step of generating summary data based on the divided time series data; and A time series data processing method for artificial intelligence comprising the step of generating a look-back window based on the above-mentioned summary data and inputting it into a predictive artificial intelligence model to predict subsequent time series data.
In claim 7, A time series data processing method for artificial intelligence, wherein the step of dividing the input time series data includes the step of dividing the input time series data at predetermined time granularities.
In claim 7, The step of generating summary data based on the aforementioned divided time series data is A step of generating segments by performing at least one of mean, median, minimum, and maximum on time series data divided for each time granule; and A time series data processing method for artificial intelligence comprising the step of generating summary data by performing normalization on the generated segment.
In Paragraph 7, The step of generating a look-back window based on the summary data generated above, inputting it into a predictive artificial intelligence model, and predicting subsequent time series data A time series data processing method for artificial intelligence, comprising the step of generating a lookback window based on the generated summary data and time series data for the summary data, inputting it into the prediction artificial intelligence model, and predicting subsequent time series data.
In Paragraph 10, A method for processing time series data for artificial intelligence, wherein the step of generating a look-back window based on the above-generated summary data and inputting it into a predictive artificial intelligence model to predict subsequent time series data includes the step of inputting the generated look-back window into a predictive artificial intelligence model utilizing at least one of a neural network, a decision tree, and a support vector machine to predict subsequent time series data.
In Paragraph 7, A time series data processing method for artificial intelligence further comprising the step of determining prediction accuracy based on real-time feedback on the predicted subsequent time series data.

Description

Time series data processing method for artificial intelligence and system supporting the same The present invention relates to a time-series data processing method for artificial intelligence and a system supporting the same. Time series data is a set of data points that are measured and recorded sequentially over time, with each measured data point recorded at specific time intervals. Such time series data can be measured in many industrial fields, including economics, finance, meteorology, healthcare, manufacturing, and energy management. Meanwhile, time series forecasting methods utilizing time series data are techniques for making decisions by predicting future data from past and present time series data. However, conventional time-series data forecasting methods have a problem in that they cannot fully utilize the temporal characteristics of the data because they require a large amount of processing resources to process large amounts of data. In particular, artificial intelligence models that require a long look-back window face the problem of increased training time and processing resource usage. FIG. 1 is a configuration diagram of a time series data processing system for artificial intelligence according to one embodiment of the present invention. FIG. 2 is an example diagram of summary data generation in a time series data processing system for artificial intelligence according to one embodiment of the present invention. FIG. 3 is an example diagram of summary data generation in a time series data processing system for artificial intelligence according to another embodiment of the present invention. FIG. 4 is an example diagram of a lookback window generation in a time series data processing system for artificial intelligence according to an embodiment of the present invention. FIG. 5 is a flowchart of a time series data processing method for artificial intelligence according to one embodiment of the present invention. Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the attached drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims. In relation to the description of the drawings, the same or corresponding components may be assigned the same reference number. Although terms such as "first," "second," etc. are used to describe various elements, components, and/or sections, it goes without saying that these elements, components, and/or sections are not limited by these terms. These terms are used merely to distinguish one element, component, or section from another. Accordingly, it goes without saying that the first element, first component, or first section mentioned below may be a second element, second component, or second section within the technical scope of the present invention. The terms used herein are for describing embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used herein, "comprises" and/or "made of" do not exclude the presence or addition of one or more other components, steps, actions, and/or elements to the mentioned components, steps, actions, and/or elements. Unless otherwise defined, all terms used in this specification (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise. Hereinafter, the configuration of the present invention will be described in detail with reference to the attached drawings. FIG. 1 is a configuration diagram of a time series data processing system for artificial intelligence according to one embodiment of the present invention. Referring to FIG. 1, a time series data processing system (100) for artificial intelligence according to one embodiment (hereinafter referred to as the artificial intelligence time series data processing system (100)) includes a data input unit (110) for receiving time series data, a data splitting unit (120) for splitting the received time series data, a summary data generation unit (130) for generating summary data based on the split time series data, and a data prediction unit (140) for generating a lookback window based on the generated summary data and inputting