Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method for performing on-demand play quantity prediction and memory scheduling on programs

A technology for storing scheduling and on-demand volume, applied in the field of communication, can solve the problems of low accuracy and low efficiency of prediction of popular programs, and achieve the effect of improving the accuracy and the efficiency of prediction

Inactive Publication Date: 2012-06-20
UNIV OF SCI & TECH OF CHINA
View PDF4 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The shortcoming of the method for predicting popular programs in the IPTV service in the above-mentioned prior art is: the accuracy and efficiency of the prediction of popular programs by this method are relatively low

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
  • Method for performing on-demand play quantity prediction and memory scheduling on programs
  • Method for performing on-demand play quantity prediction and memory scheduling on programs
  • Method for performing on-demand play quantity prediction and memory scheduling on programs

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0024] The neural network is also called the error backpropagation neural network. It not only has input nodes and output nodes, but also has more than one layer of hidden nodes. In theory, a three-layer BP with only one hidden layer can approximate any Boolean function. Because the neural network has the advantages of good nonlinear function approximation ability and fast training speed, it can achieve good results by using the neural network to predict the amount of on-demand programs.

[0025] The processing flow of the method for predicting the amount of on-demand programs based on the neural network provided by this embodiment is as follows: figure 1 As shown, the following processing steps are included:

[0026] Step 11. Determine input and output data for training the neural network, and preprocess the input data.

[0027] For hot program prediction, since there are only two types of predicted programs, hot or not, the output data of the neural network in the embodimen...

Embodiment 2

[0071] Based on the above-mentioned predicted on-demand volume of each program in the current day, the processing flow of a method for storing and scheduling programs among multiple nodes provided by this embodiment is as follows: figure 2 shown, including the following steps:

[0072] Step 21: Layer the selected programs according to the on-demand volume of the programs, and store the programs of each layer in a differentiated manner on the central node and each edge node according to a predetermined storage strategy.

[0073] Sort all the programs from high to low according to the amount of demand, select the top N programs according to the demand, and divide the N programs into L layers according to the amount of demand, and the threshold of each layer is: T 1 >T 2 >T 3 >...>T L , the number of programs stored in each layer corresponds to: H 1 , H 2 , H 3 ,...H L . When the above threshold T 1 ,T 2 ,T 3 ,T L When choosing different values, H 1 , H 2 , H 3 ,....

Embodiment 3

[0085] This embodiment provides a device for forecasting the amount of on-demand and storing and scheduling programs, and its specific structure is as follows: image 3 shown, including:

[0086] The input data acquisition module 31 is used to obtain the input data of the neural network according to the on-demand volume of each program in the previous setting number days of the designated day;

[0087] The training processing module 32 is used to train the pre-established neural network according to the input data acquired by the input data acquisition module and the on-demand volume of each program in the specified day;

[0088] The prediction processing module 33 is used to input the on-demand volume of each program in the previous set number of days of the current day into the neural network after training as input data, and use the output data obtained by calculating the neural network after the training as prediction The on-demand volume of each program in the current da...

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 embodiment of the invention provides a method for performing on-demand play quantity prediction and memory scheduling on programs, which comprises the following steps: acquiring input data of a neural network according to on-demand play quantities of every program in a number of days preset before specific days; training the previously established neural network according to the input data and the on-demand play quantities of every program within the specific days; and acquiring the input data of the trained neural network according to the on-demand play quantities of every program within the number of days preset before the current day, inputting the input data into the trained neural network, and using the computed output data of the trained neural network as the predicted on-demand play quantities of every program within the current day. According to the embodiment of the invention, the on-demand play quantities of the program is predicted by using the neural network, thereby acquiring the heat of every program; and the three-layer neural network approaches to the characteristics of any Boolean function, thereby enhancing the prediction accuracy and prediction efficiency of hot programs.

Description

technical field [0001] The invention relates to the field of communication technology, in particular to a method for forecasting the amount of on-demand programs and storing and scheduling programs. Background technique [0002] With the development of network technology, the scale of global broadband users is increasing day by day, and more and more information obtained by people on the broadband IP network appears in formats such as images, sounds, and videos. Digital technology has greatly facilitated the storage, transmission and duplication of video information, making the amount of digital video content grow rapidly. The video multimedia on demand service represented by VOD (Video On Demand, video on demand service) is rapidly becoming an important application of the Internet because of its interactivity and immediacy. [0003] IPTV (Internet Protocol, Internet TV) uses a broadband IP network to provide a variety of digital media services including digital TV programs...

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
IPC IPC(8): H04N21/258G06F17/30G06N3/02
Inventor 李俊洪爽万敏马书超
Owner UNIV OF SCI & TECH OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com