A Publish and Subscribe Method Based on Historical Record Data Mining
A technology of historical records and data mining, applied in network data retrieval, electronic digital data processing, other database retrieval, etc., can solve the problem of prolonged publishing to reaching subscribers, not considering the coverage relationship between subscriptions, poor load balancing capabilities, etc. problem, to achieve the effect of simplicity, overall performance improvement, and improved relevance
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment
[0113] A simulation system is built on the OverSim simulation platform, and the algorithm in the present invention (publish-subscribe system based on data mining, Data Mining based Publish / Subscribe System, SMPSS) is compared with Ferry and Eferry.
[0114] There are seven attributes in this publish / subscribe system, which are defined as:
[0115] S={[Date:string,2 / Jan / 98,31 / Dec / 02],
[0116] [Symbol: string,"aaa","zzz"],
[0117] [Open:float,0,500],
[0118] [Close:float,0,500],
[0119] [High:float,0,500],
[0120] [Low:float,0,500],
[0121] [Volume: integer, 0, 310000000]}.
[0122] There are 6 subscription generation templates, which are:
[0123] T 1 ={(Symbol=P 1 )∩(P 2 ≤Open≤P 3 )} with a probability of 20%;
[0124] T 2 ={(Symbol=P 1 )∩(Low≤P 2 )} with a probability of 25%;
[0125] T 3 ={(Symbol=P 1 )∩(High≥P 2 )} with a probability of 30%;
[0126] T 4 ={(Symbol=P 1 )∩(Volume≥P 2 )} with a probability of 10%;
[0127] T 5 ={(Volume≥P 1 )} with...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


