Causal modeling method and device for event sequence

A technology of event sequence and modeling method, applied in special data processing applications, instruments, electrical digital data processing, etc., can solve the problem that the causal detection method cannot be directly applied to discrete event sequences, etc., and achieve quality assurance, simple implementation method, The effect of flexibility

Pending Publication Date: 2020-12-11
ZHEJIANG LAB
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiments of the present invention is to provide an event sequence-oriented causal modeling method and device to solve the problem that existing causal detection methods cannot be directly applied to discrete event sequences

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Causal modeling method and device for event sequence
  • Causal modeling method and device for event sequence
  • Causal modeling method and device for event sequence

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] figure 1 It is a flow chart of an event-sequence-oriented causal modeling method provided in an embodiment of the present invention; a kind of event-sequence-oriented causal modeling method provided in this embodiment includes the following steps:

[0027] Step S101, obtaining multiple pieces of event sequence data, wherein each event sequence S includes multiple events with a sequence relationship, and the events in the event sequence all come from a shared event set;

[0028] The above-mentioned events may be air pollution transmission or sports events, and of course are not limited to the two types listed here.

[0029] Step S102, converting the event sequence data into a vectorized representation, and completing the clustering of the event sequence;

[0030] Wherein the vectorization method adopts a Bag-of-Word (Bag-of-Word) model, and the number of occurrences of each event in the sequence is counted to complete the vectorization. The clustering method adopts an ...

Embodiment 2

[0041] The present invention is based on a causal modeling method oriented to event sequences, which comprises the following steps when applied to the analysis of the transmission source of air pollution:

[0042] Step 1: Obtain the air pollution event sequence S related to A according to the target city A and the surrounding city set C;

[0043] Specifically, obtain the target city A and the set of surrounding cities C, and construct an event sequence related to A’s air pollution, where the event definition consists of cities and pollution levels, such as C 1 - Mild, C 2 -Heavy (there are two levels of pollution, i.e. light pollution and heavy pollution); for each air pollution event A-mild or A-heavy (defined as the target event) of A to find the preceding event and construct a sequence S, The standard of the preceding event is that the event is a pollution event that occurs in the city set of C, and the occurrence time is not earlier than 12 hours of the target event, and ...

Embodiment 3

[0060] The present invention is based on a kind of event-sequence-oriented causal modeling method, and comprises the following steps when being applied to the motion event sequence of analyzing table tennis:

[0061] Step 1: According to the existing definition of table tennis skills, convert each shot of the players in each round into the corresponding table tennis skills (such as kicking, smashing, arcing, etc.), and then convert the two players in each round Swing records are transformed into sequences of table tennis techniques, and finally a collection of event sequences is obtained.

[0062] Step 2: Transform the table tennis technology sequence into a vectorized representation, and complete the clustering of the sequence on this basis: the vectorization method uses the Bag-of-Word model to count each table tennis ball in the sequence The number of occurrences of the technology is vectorized; the clustering method adopts the existing density-based clustering algorithm. ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a causal modeling method and device for event sequences, and the method comprises the steps: obtaining a plurality of pieces of event sequence data, wherein each event sequencecomprises a plurality of events with the sequential relation, and the events in the event sequences come from a shared event set; converting the event sequence data into vectorized representation, and completing clustering of event sequences; and acquiring a causal network corresponding to each group of event sequences in the clustering result by using a time-preserving detection method, whereineach node in the causal network is an event, and the causal relationship between the nodes is represented through directed edge connection. According to the invention, a new causal detection frameworkis created, the event sequence-oriented causal detection method is realized, the realization method is simple and convenient, the means is flexible, and the quality of the causal relationship can beremarkably guaranteed.

Description

technical field [0001] The invention relates to the field of computer causal analysis and visualization, in particular to an event sequence-oriented causal modeling method and device. Background technique [0002] Causal analysis of event sequence data can characterize the relationship between events and can play an important role in various fields, such as marketing behavior analysis, electronic medical records and healthcare analysis, and error log analysis, etc. Controlled experiments are a common way to deduce the cause of events, but due to the high cost of experimental setup, controlled experiments are not suitable for the application field in many cases. Based on this, many research works use co-occurrence relations to model the connection between events. Co-occurrence-based methods assume that frequently co-occurring events are highly correlated. However, co-occurrence is not causation and the insights provided are ambiguous. Therefore, it remains essential to cha...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/2458
CPCG06F16/2474
Inventor 巫英才谢潇何墨琪
Owner ZHEJIANG LAB
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap