Time Gram Encoding System for Event Time Series Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for encoding event time series fail to consider inter-arrival times between events, resulting in incomplete pattern recognition and failure to capture temporal dimensions in data analysis.

Innovation Solution

A system that computes inter-arrival times, transforms them into discrete symbols, and maps these symbols using a dictionary to generate 'time grams,' which encode both event sequences and time gaps, enabling more comprehensive pattern recognition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If conventional methods are used to encode event time series, then the encoding process is simple and fast, but the temporal dimension and inter-arrival times between events are not captured

Engineering Contradiction:
Improvetemporal informationVSAvoidencoding system complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The encoding system segments the continuous inter-arrival time into discrete time symbols by dividing the time range into multiple bins. Each bin represents a discrete time symbol, transforming continuous temporal information into discrete categories that can be processed while preserving the temporal dimension of event sequences

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary transformation process that converts continuous inter-arrival times into discrete time symbols through a dictionary mapping. This intermediary step bridges the gap between continuous temporal data and discrete event sequences, enabling temporal information to be preserved in the encoded representation

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If inter-arrival times are transformed into discrete symbols, then pattern recognition accuracy is improved, but the processing complexity increases

Engineering Contradiction:
Improvepattern recognition accuracyVSAvoidtransformation process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system changes the parameter representation of inter-arrival times from continuous values to discrete symbols by applying a binning transformation. This parameter change enables more precise pattern recognition by preserving temporal distinctions while making the data suitable for discrete sequence processing algorithms

Inventive Principle:
Principle #35Parameter changes

3Loss of information

If a dictionary mapping is used to encode events and time symbols, then the temporal patterns are preserved, but the encoding complexity increases

Engineering Contradiction:
Improvetemporal pattern informationVSAvoidmapping process complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system performs preliminary action by pre-defining a dictionary that maps event types and time symbols to discrete representations before the actual encoding process. This pre-prepared mapping structure enables efficient encoding while preserving temporal patterns, as the transformation rules are established in advance rather than computed in real-time

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10049140B2Encoding system, method, and recording medium for time grams
Publication Date: 2018.08.14 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10049140B2 patent drawing
  • US10049140B2 patent drawing
  • US10049140B2 patent drawing

AI summary

An encoding system for encoding an event time series, the system including an inter-arrival time computing device configured to compute an inter-arrival time between a plurality of input events and computes a sequence of events, a transformation device configured to transform the inter-arrival time between the plurality of input events into discrete time symbols, and a mapping device configured to map the input events and the discrete time symbols using a dictionary to output a time gram.