CN113812966A

CN113812966A - Method and device for preference prediction based on multimedia information

Info

Publication number: CN113812966A
Application number: CN202110968122.XA
Authority: CN
Inventors: 刘光远; 岳远昊; 周巍
Original assignee: Southwest University
Current assignee: Southwest University
Priority date: 2021-08-23
Filing date: 2021-08-23
Publication date: 2021-12-21

Abstract

Embodiments of the present application provide a method and device for preference prediction based on multimedia information, which relate to the field of EEG. The method for preference prediction based on multimedia information includes: acquiring EEG signals to be identified according to preset multimedia information; Identifying the EEG signal and performing preprocessing to obtain a preprocessed EEG signal; selecting from the preprocessed EEG signal to obtain an effective EEG signal; performing feature extraction on the effective EEG signal to obtain an EEG feature; The EEG feature is predicted to obtain a preference prediction result corresponding to the multimedia information. It can be seen that, by implementing this embodiment, people's real preferences can be accurately fed back through EEG signals, so as to know people's real preferences.

Description

Preference prediction method and device based on multimedia information

Technical Field

The application relates to the field of electroencephalogram, in particular to a preference prediction method and device based on multimedia information.

Background

Distributors want to know their preferences, whether they are film or music works. The most common way is to try on and listen. However, the results obtained in this manner often rely on the consumer's self-statement. Due to various reasons, the method can cause that people cannot accurately express their preferences, so that the real preferences of people cannot be known.

Disclosure of Invention

The embodiment of the application aims to provide a preference prediction method and device based on multimedia information, which can accurately feed back the real preference of people through electroencephalogram signals, so that the real preference condition of people can be known.

A first aspect of an embodiment of the present application provides a method for predicting a preference based on multimedia information, where the method includes:

acquiring an electroencephalogram signal to be identified according to preset multimedia information;

preprocessing the electroencephalogram signal to be identified to obtain a preprocessed electroencephalogram signal;

selecting the preprocessed electroencephalogram signals to obtain effective electroencephalogram signals;

extracting the characteristics of the effective electroencephalogram signals to obtain electroencephalogram characteristics;

and predicting the electroencephalogram characteristics to obtain a preference prediction result corresponding to the multimedia information.

In the implementation process, the method can play preset multimedia information for a testee and simultaneously acquire the electroencephalogram signals to be identified; after acquiring the electroencephalogram signal to be identified, preprocessing the electroencephalogram signal to be identified, selecting the processed effective electroencephalogram signal for feature extraction processing, and then performing preference prediction on the extracted electroencephalogram feature to obtain a preference prediction result corresponding to the multimedia information. Therefore, by implementing the implementation mode, the real preference of people can be fed back through the electroencephalogram signals based on the multimedia information of the testee, and therefore the real preference condition of people can be accurately obtained.

Further, the step of acquiring the electroencephalogram signal to be identified according to the preset multimedia information comprises the following steps:

outputting preset multimedia information to a testee;

and acquiring the electroencephalogram signals to be identified, which are generated when the testee perceives the multimedia information.

In the implementation process, the method can acquire the electroencephalogram signals of the testee while outputting the multimedia information, so that the real brain reaction of the testee is acquired, the real preference of people is accurately fed back, and the real preference condition of people is known.

Further, the step of preprocessing the electroencephalogram signal to be identified to obtain a preprocessed electroencephalogram signal comprises the following steps:

and carrying out segmentation processing, filtering processing, artifact removing processing and interpolation bad lead processing on the electroencephalogram signal to be identified to obtain a preprocessed electroencephalogram signal.

In the implementation process, the electroencephalogram signals to be identified can be preprocessed to obtain the electroencephalogram signals which can be used for screening, so that useless electroencephalogram signals are prevented from being identified and extracted, the preference prediction efficiency is greatly improved, and meanwhile, the accuracy of preference prediction can be guaranteed.

Further, the step of selecting from the preprocessed electroencephalogram signals to obtain effective electroencephalogram signals comprises:

and selecting effective electroencephalogram signals from the preprocessed electroencephalogram signals according to the main correlation value of each channel corresponding to the preprocessed electroencephalogram signals.

In the implementation process, effective electroencephalogram channels can be determined in 128 channels according to the main body correlation value, and effective electroencephalogram signals can be acquired. Therefore, by adopting the implementation mode, the acquired electroencephalogram signals can be promoted to be effective by selecting the effective electroencephalogram channels, so that the prediction efficiency is greatly improved, and the preference prediction accuracy can be improved.

selecting a prefrontal lobe electroencephalogram signal channel and a temporal lobe electroencephalogram signal channel from the preprocessed electroencephalogram signals;

and determining beta-frequency-range electroencephalogram signals and delta-frequency-range electroencephalogram signals separated from the prefrontal-lobe electroencephalogram signal channel and the temporal-lobe electroencephalogram signal channel as effective electroencephalogram signals.

In the implementation process, the method selects the electroencephalogram signals of a specific electroencephalogram signal channel, and extracts the most appropriate electroencephalogram characteristics according to the electroencephalogram signals. Therefore, the selection process of the electroencephalogram signal channel can be avoided, and the preference prediction efficiency can be further improved.

Further, the step of extracting the features of the effective electroencephalogram signal to obtain the electroencephalogram features comprises the following steps:

and performing feature extraction on the effective electroencephalogram signal according to a Fourier transform algorithm to obtain electroencephalogram features including power spectral density features and differential entropy features.

In the implementation process, more effective electroencephalogram characteristics can be acquired, so that a powerful prediction basis is provided for subsequent preference prediction.

Further, the step of predicting the electroencephalogram characteristics to obtain a preference prediction result corresponding to the multimedia information includes:

and predicting the electroencephalogram characteristics according to a gradient rising decision tree algorithm to obtain a preference prediction result corresponding to the multimedia information.

In the implementation process, the prediction of the electroencephalogram characteristics can be completed through artificial intelligence, so that a more accurate preference prediction result is obtained, and the real preference condition of people can be fed back.

A second aspect of the embodiments of the present application provides a multimedia information-based preference prediction apparatus, including:

the acquisition unit is used for acquiring the electroencephalogram signals to be identified according to preset multimedia information;

the preprocessing unit is used for preprocessing the electroencephalogram signal to be identified to obtain a preprocessed electroencephalogram signal;

the selection unit is used for selecting from the preprocessed electroencephalogram signals to obtain effective electroencephalogram signals;

the extraction unit is used for carrying out feature extraction on the effective electroencephalogram signal to obtain electroencephalogram features;

and the prediction unit is used for predicting the electroencephalogram characteristics to obtain a preference prediction result corresponding to the multimedia information.

In the implementation process, the device can feed back the real preference of people through the electroencephalogram signal based on the multimedia information of the testee, so that the real preference condition of people can be accurately known.

A third aspect of the embodiments of the present application provides an electronic device, including a memory and a processor, where the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the method for predicting multimedia-information-based preference according to any one of the first aspect of the embodiments of the present application.

A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, which stores computer program instructions, and when the computer program instructions are read and executed by a processor, the computer program instructions perform the method for multimedia information based preference prediction according to any one of the first aspect of the embodiments of the present application.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.

Fig. 1 is a flowchart illustrating a method for predicting a preference based on multimedia information according to an embodiment of the present disclosure;

FIG. 2 is a flowchart illustrating another method for predicting multimedia-information-based preferences according to an embodiment of the present disclosure;

FIG. 3 is a schematic structural diagram of an apparatus for predicting multimedia-information-based preferences according to an embodiment of the present disclosure;

FIG. 4 is a schematic structural diagram of another multimedia information-based preference prediction apparatus according to an embodiment of the present application;

fig. 5 is a first half of an exemplary flowchart of a method for predicting multimedia-information-based preferences according to an embodiment of the present disclosure;

fig. 6 is a second half of an exemplary flowchart of a method for predicting a preference based on multimedia information according to an embodiment of the present disclosure;

FIG. 7 is a schematic diagram illustrating electroencephalogram signal channel selection provided by an embodiment of the present application;

FIG. 8 is a diagram illustrating the variation of ISC value with playing time of a movie trailer according to an embodiment of the present application;

fig. 9 is a diagram of creating an emotion model according to an embodiment of the present application;

fig. 10 is a schematic flow chart of prediction of suitable music for adjusting emotion according to an embodiment of the present application;

fig. 11 is a schematic diagram of a three-stage experimental process of the emotion model shown in fig. 9 according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.

Example 1

Referring to fig. 1, fig. 1 is a flowchart illustrating a preference prediction method based on multimedia information according to an embodiment of the present disclosure. The preference prediction method based on the multimedia information comprises the following steps:

s101, acquiring the electroencephalogram signals to be identified according to preset multimedia information.

As an optional implementation manner, before the step of acquiring the electroencephalogram signal to be identified according to preset multimedia information, the method may further include:

predicting an experimental paradigm according to preset preference to acquire electroencephalograms to be recognized; the preference prediction experiment paradigm comprises multimedia information corresponding to a plurality of emotion types and a plurality of emotion intensities;

preprocessing an electroencephalogram signal to be identified to obtain a preprocessed electroencephalogram signal;

predicting the electroencephalogram characteristics to obtain a preference prediction result corresponding to the preference prediction experiment paradigm;

and determining the most appropriate electroencephalogram signal channel in the 128 electroencephalogram signal channels according to the preference prediction result.

In this embodiment, the most suitable electroencephalogram signal channel is used for direct selection in step S103, so that the selection process of the electroencephalogram signal channel can be avoided, and the prediction efficiency can be improved.

As a further optional implementation, generating a multimedia information preference prediction model according to the preference prediction result corresponding to the preference prediction experimental paradigm and the preference prediction experimental paradigm; so that the method can predict the preference of the testee for the multimedia information through the multimedia information preference prediction model.

In this embodiment, the method allows each subject to view 12 movie trailers (movie preference design paradigm) in the experiment. Before watching each movie trailer, prompting the subject which emotion type the trailer to be played belongs to; and after the playing of each trailer is finished, the following two problems are presented to the testee: (1) how much emotional intensity is; (2) how much will it want to watch the complete movie further through the movie trailer. In this process, the subject was asked to rest for 3 minutes after each 3-movie trailer. Therefore, the method can acquire the electroencephalogram channels required to be selected by the multimedia preference prediction model by designing an experimental paradigm.

For example, the method can pre-design a film preference design paradigm, then play the film preference design paradigm for a testee, and simultaneously acquire electroencephalogram signals; after acquiring an electroencephalogram signal, carrying out preprocessing such as segmentation, re-reference, filtering, eye charge removal, interpolation bad conduction and the like on the electroencephalogram signal; then, selecting a time period of an electroencephalogram channel and an electroencephalogram signal by using main body correlation (ISC, after the ISC is subjected to the same natural stimulation, similar neural responses can be caused in a plurality of audiences and are called as main body correlation), and obtaining a selected effective electroencephalogram signal; then, performing feature extraction on the selected effective electroencephalogram signal by adopting a discrete Fourier transform algorithm to obtain electroencephalogram features including power spectral density features (PSD) and differential entropy features (DE) of the effective electroencephalogram signal; and finally, predicting the electroencephalogram characteristics by adopting a gradient ascending decision tree algorithm (GBDT) to obtain a movie preference prediction result of the electroencephalogram signals. Therefore, the method can obtain a plurality of film preference prediction results by using a plurality of film preference design paradigms, and optimize and adjust the process according to actual information to obtain a multimedia information preference prediction model. Thereby enabling the multimedia information preference prediction model to implement the process described by the method.

By implementing the implementation mode, the EEG channel and the EEG signal time period can be selected through the ISC, so that the selected EEG signal can well show the interest intensity of different viewers, the extracted power spectral density and differential entropy characteristics can well reflect emotional changes, and a regression model established through a gradient ascent decision tree algorithm obtains a good prediction effect on a data set.

S102, preprocessing the electroencephalogram signal to be recognized to obtain a preprocessed electroencephalogram signal.

In this embodiment, the method for preprocessing the electroencephalogram signal to be identified is not limited at all.

S103, selecting the pre-processed electroencephalogram signals to obtain effective electroencephalogram signals.

In the embodiment, an effective electroencephalogram signal channel is preferentially selected from the preprocessed electroencephalogram signals, and then the effective electroencephalogram signals in the effective electroencephalogram signal channel are obtained.

In the embodiment, the electroencephalogram signal channel of the electroencephalogram signal to be selected can be predetermined through experiments, so that the preference prediction efficiency can be effectively improved.

And S104, performing feature extraction on the effective electroencephalogram signals to obtain electroencephalogram features.

And S105, predicting the electroencephalogram characteristics to obtain a preference prediction result corresponding to the multimedia information.

In this embodiment, the preference prediction result is used to indicate the interest strength of the multimedia information by different viewers.

As an optional implementation manner, the step of predicting the electroencephalogram characteristics to obtain a preference prediction result corresponding to the multimedia information includes:

and inputting the electroencephalogram characteristics into the multimedia information preference prediction model, and acquiring a preference prediction result which is output by the multimedia information preference prediction model and corresponds to the multimedia information.

In the embodiment of the present application, the execution subject of the method may be a computing device such as a computer and a server, and is not limited in this embodiment.

In this embodiment, an execution subject of the method may also be an intelligent device such as a smart phone and a tablet computer, which is not limited in this embodiment.

Therefore, by implementing the preference prediction method based on multimedia information described in the embodiment, the electroencephalogram signal to be identified can be acquired while preset multimedia information is played for a subject; after acquiring the electroencephalogram signal to be identified, preprocessing the electroencephalogram signal to be identified, selecting the processed effective electroencephalogram signal for feature extraction processing, and then performing preference prediction on the extracted electroencephalogram feature to obtain a preference prediction result corresponding to the multimedia information. Therefore, by implementing the implementation mode, the real preference of people can be fed back through the electroencephalogram signals based on the multimedia information of the testee, and therefore the real preference condition of people can be accurately obtained.

Example 2

Referring to fig. 2, fig. 2 is a flowchart illustrating a method for predicting multimedia information-based preferences according to an embodiment of the present application. As shown in fig. 2, wherein the multimedia information-based preference prediction method includes:

s201, outputting preset multimedia information to a testee.

S202, collecting electroencephalogram signals to be identified, which are generated when a testee perceives multimedia information.

In the embodiment of the application, by implementing the steps S201 to S202, the electroencephalogram signal to be identified can be acquired according to preset multimedia information.

S203, carrying out segmentation processing, filtering processing, artifact removing processing and interpolation bad lead processing on the electroencephalogram signal to be identified to obtain a preprocessed electroencephalogram signal.

In the embodiment of the application, the step S203 is implemented, so that the electroencephalogram signal to be recognized can be preprocessed, and the preprocessed electroencephalogram signal can be obtained.

After step S203, the following steps are also included:

s204, selecting the preprocessed electroencephalogram signals to obtain effective electroencephalogram signals.

As an optional implementation manner, selecting from the preprocessed electroencephalogram signals to obtain an effective electroencephalogram signal may include the following steps:

Referring to fig. 7, fig. 7 is a schematic diagram illustrating electroencephalogram channel selection. As can be seen, the method performs a correlation component analysis on 30 subject data of one video stimulus, respectively, to obtain a subject correlation value (ISC) for each channel. Subsequently, the 128 channels are arranged in descending order of the subject relevance values. After the channels with the main body correlation smaller than zero are removed, the main body correlation values of the rest channels are mostly concentrated between 0.0 and 0.1. Therefore, the threshold value is set to 0.1 herein. Finally, only channels with a subject correlation greater than 0.1 are selected. The above operation was performed for all 12 video stimuli, and a channel selection result was obtained. Since all stimuli require a uniform data analysis, the final defined channels are common among the results of the 12 video stimuli, the selected channels being a1, a2, A3, a4, a5, a 6.

Referring to fig. 8, fig. 8 is a diagram illustrating a variation of the ISC value according to the playing time of a trailer. For each video stimulus, the method calculates the subject correlation for each sliding window on the a1, a2, A3, a4, a5, a6 channels by using a sliding window of 5s duration with once-a-second window movement. Wherein, fig. 8 shows the variation curve of ISC of 5s time period on A3 channel in movie3 with time, and the method selects 5s movie trailer time period of ISC peak value on each channel as the brain signal time period.

and determining beta-frequency-range electroencephalogram signals and delta-frequency-range electroencephalogram signals separated from a prefrontal-lobe electroencephalogram signal channel and a temporal-lobe electroencephalogram signal channel as effective electroencephalogram signals.

Referring to fig. 9, fig. 9 is a diagram of creating an emotion model. As shown in fig. 9, the emotion model is the multimedia information preference prediction model described above. The model may be used for preference prediction for multimedia information.

Referring to fig. 10, fig. 10 is a schematic diagram illustrating a process of predicting suitable music for adjusting mood. In this figure, the emotion model shown in fig. 9 can select appropriate music (positive music or negative music) to play according to the emotion of the subject, thereby achieving the effect of emotion adjustment on the subject.

In the embodiment, the mode of adjusting the emotion by music is mainly to adjust the emotion of a person by generating emotional resonance of the person or by actions such as memory caused by music, and researches indicate that negative emotion of the person can be relieved by emotional resonance of sad music in the process of adjusting the emotion of the person, and negative music in the embodiment has the beneficial effect that the emotion can be effectively restored to a calm state from a tense excited state in the aspect of emotional arousal degree.

Referring to fig. 11, fig. 11 is a schematic diagram illustrating a three-stage experimental process based on the emotion model. Wherein, the three stages of the experiment are respectively: an emotion multi-sample collection stage; negative music regulates the negative mood phase; positive music regulates the negative emotional phase. Specifically, the data of the first stage is used to establish an emotion model, which is used to predict the emotional states of the second and third stages from two dimensions (value and arousal) of emotion, and further to infer the type of music that the user effectively adjusts his own emotion.

The specific experimental procedure is illustrated in detail by way of example:

stage one: the aim of the design of the stage is to establish an emotion model about a testee, and the emotion scores, namely the arousal degree and the valence score, of the testee in the later two stages are predicted according to the emotion model based on the testee, so that the research on the similarity and difference comparison of the two types of music on emotion regulation is facilitated. In this section, the subject will watch five emotional videos, which all ensure that the emotion is sufficiently aroused and the strength of the arousal is also guaranteed. Two of the five videos are positive, two negative, and one neutral. After each video is played, they will give a rating and arousal score according to their mood. The electroencephalogram data collected in the process are used for emotion modeling, and the electroencephalogram data are used as emotion labels according to emotion scores given by the emotion modeling. After the first experiment stage is finished, the testee takes a rest, and when the emotional state is recovered to a neutral emotion and calm state, the testee actively enters an experiment stage two.

And a second stage: negative music regulates the negative emotional experimental phase. The purpose of the experiment at this stage was to investigate the effect of negative music in regulating negative mood. The negative emotion of the testee, namely the emotion of the sad negative face type, is initially aroused by utilizing a piece of negative video. After the video is played, the negative music playing stage is started in time, and the testee is instructed to sit still and enjoy music seriously during the music playing process. The data set collected at this stage will be used as a test set for emotion model testing to determine if there is a significant improvement in regulating negative emotions through negative music. Also, at this stage, the subjects will give their emotional scores, whether after the music or video has been played. After the end of this phase, the emotional state of the subject was restored to calm and neutral mood, and the third phase of the experiment was entered.

And a third stage: positive music regulates the negative emotional experimental phase. At this stage, a negative video is also needed to arouse the negative emotion of the testee, and after the video is played, positive music, namely positive music, is used as a regulating tool for the negative emotion. The electroencephalogram data collected at the stage are used as a test set in a prediction model, and the emotion difference before and after emotion adjustment is determined according to the test result.

Wherein the emotional state comprises a valence and an arousal score of the emotion, the score is 1 to 9 points, the valence (valance) is expressed as unpleasant to very pleasant, 5 points are neutral emotions, the arousal (arousal) is expressed as very calm to very excited or tense, and 5 points are expressed as a calm to excited transition line.

S205, extracting the characteristics of the effective electroencephalogram signals according to a Fourier transform algorithm to obtain electroencephalogram characteristics including power spectral density characteristics and differential entropy characteristics.

In the embodiment of the application, by implementing the step S205, feature extraction can be performed on the effective electroencephalogram signal to obtain electroencephalogram features.

S206, predicting the electroencephalogram characteristics according to a gradient ascending decision tree algorithm to obtain a preference prediction result corresponding to the multimedia information.

In the embodiment of the present application, by implementing the step S206, the electroencephalogram characteristics can be predicted, and a preference prediction result corresponding to the multimedia information is obtained.

Referring to fig. 5 and 6, fig. 5 and 6 together show a flow chart of a method for predicting movie preferences. Fig. 5 and 6 illustrate the flow of the method accurately.

Therefore, the preference prediction method based on the multimedia information described in the embodiment can feed back the real preference of people through the electroencephalogram signal based on the multimedia information of the testee, so that the real preference condition of people can be accurately known.

Example 3

Referring to fig. 3, fig. 3 is a schematic structural diagram of a multimedia information-based preference prediction apparatus according to an embodiment of the present application. As shown in fig. 3, the multimedia information-based preference prediction apparatus includes:

the acquiring unit 310 is used for acquiring an electroencephalogram signal to be identified according to preset multimedia information;

the preprocessing unit 320 is used for preprocessing the electroencephalogram signal to be identified to obtain a preprocessed electroencephalogram signal;

the selection unit 330 is used for selecting the preprocessed electroencephalogram signals to obtain effective electroencephalogram signals;

the extraction unit 340 is configured to perform feature extraction on the effective electroencephalogram signal to obtain electroencephalogram features;

and the prediction unit 350 is configured to predict the electroencephalogram characteristics to obtain a preference prediction result corresponding to the multimedia information.

In the embodiment of the present application, for the explanation of the preference prediction apparatus based on multimedia information, reference may be made to the description in embodiment 1 or embodiment 2, and details are not repeated in this embodiment.

Therefore, the preference prediction device based on the multimedia information described in the embodiment can feed back the real preference of people through the electroencephalogram signal based on the multimedia information of the testee, so that the real preference condition of people can be accurately known.

Example 4

Referring to fig. 4, fig. 4 is a schematic structural diagram of a multimedia information-based preference prediction apparatus according to an embodiment of the present disclosure. Among them, the multimedia information based preference prediction apparatus shown in fig. 4 is optimized by the multimedia information based preference prediction apparatus shown in fig. 3. As shown in fig. 4, the obtaining unit 310 includes:

an output subunit 311, configured to output preset multimedia information to the subject;

and the collecting subunit 312 is configured to collect an electroencephalogram signal to be identified, which is generated when the subject perceives the multimedia information.

As an optional implementation manner, the preprocessing unit 320 is specifically configured to perform segmentation processing, filtering processing, artifact removal processing, and interpolation bad-lead processing on the electroencephalogram signal to be identified, so as to obtain a preprocessed electroencephalogram signal.

As an optional implementation manner, the selecting unit 330 is specifically configured to select an effective electroencephalogram signal from the preprocessed electroencephalogram signals according to the main correlation value of each channel corresponding to the preprocessed electroencephalogram signals.

As another optional implementation, the selecting unit 330 is specifically configured to select a prefrontal electroencephalogram signal channel and a temporal electroencephalogram signal channel from the preprocessed electroencephalogram signals; and determining beta-frequency-band electroencephalogram signals and delta-frequency-band electroencephalogram signals separated from the prefrontal-lobe electroencephalogram signal channel and the temporal-lobe electroencephalogram signal channel as effective electroencephalogram signals.

As an optional implementation manner, the extraction unit 340 is specifically configured to perform feature extraction on the effective electroencephalogram signal according to a fourier transform algorithm, so as to obtain electroencephalogram features including a power spectral density feature and a differential entropy feature.

As an optional implementation manner, the prediction unit 350 is specifically configured to predict the electroencephalogram characteristics according to a gradient-ascending decision tree algorithm, so as to obtain a preference prediction result corresponding to the multimedia information.

An embodiment of the present application provides an electronic device, which includes a memory and a processor, where the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the multimedia information based preference prediction method according to any one of embodiment 1 or embodiment 2 of the present application.

An embodiment of the present application provides a computer-readable storage medium, which stores computer program instructions, and when the computer program instructions are read and executed by a processor, the method for predicting multimedia information-based preference according to any one of embodiment 1 or embodiment 2 of the present application is performed.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

Claims

1. A preference prediction method based on multimedia information, wherein the method comprises:

Obtain the EEG signal to be identified according to the preset multimedia information;

Preprocessing the to-be-identified EEG signal to obtain a preprocessed EEG signal;

Selecting from the preprocessed EEG signals to obtain effective EEG signals;

Perform feature extraction on the effective EEG signal to obtain EEG features;

Predicting the EEG feature to obtain a preference prediction result corresponding to the multimedia information.

2. The preference prediction method based on multimedia information according to claim 1, wherein the step of obtaining the EEG signal to be identified according to preset multimedia information comprises:

output preset multimedia information to the subjects;

The electroencephalogram signals to be identified generated when the subject perceives the multimedia information are collected.

3. The preference prediction method based on multimedia information according to claim 1, wherein the step of preprocessing the to-be-recognized EEG signal to obtain the preprocessed EEG signal comprises:

Perform segmentation processing, filtering processing, artifact removal processing and interpolation bad derivation processing on the to-be-identified EEG signal to obtain a pre-processed EEG signal.

4. The preference prediction method based on multimedia information according to claim 1, wherein the step of selecting from the preprocessed EEG signals to obtain an effective EEG signal comprises:

According to the subject correlation value of each channel corresponding to the preprocessed electroencephalographic signal, an effective electroencephalographic signal is selected from the preprocessed electroencephalographic signal.

5. The preference prediction method based on multimedia information according to claim 1, wherein the step of selecting from the preprocessed EEG signals to obtain an effective EEG signal comprises:

Selecting a prefrontal lobe EEG signal channel and a temporal lobe EEG signal channel from the preprocessed EEG signal;

The β-band EEG signal and the δ-band EEG signal separated from the two channels of the prefrontal lobe EEG signal channel and the temporal lobe EEG signal channel are determined as effective EEG signals.

6. The preference prediction method based on multimedia information according to claim 1, wherein the step of performing feature extraction on the effective EEG signal to obtain the EEG feature comprises:

Feature extraction is performed on the effective EEG signal according to the Fourier transform algorithm to obtain EEG features including power spectral density features and differential entropy features.

7. The preference prediction method based on multimedia information according to claim 1, wherein the step of predicting the EEG feature to obtain a preference prediction result corresponding to the multimedia information comprises:

The EEG feature is predicted according to the gradient ascent decision tree algorithm, and a preference prediction result corresponding to the multimedia information is obtained.

8. A preference prediction device based on multimedia information, wherein the preference prediction device based on multimedia information comprises:

an acquisition unit, configured to acquire the EEG signal to be identified according to preset multimedia information;

a preprocessing unit, configured to preprocess the to-be-identified EEG signal to obtain a preprocessed EEG signal;

a selection unit for selecting from the preprocessed EEG signal to obtain an effective EEG signal;

an extraction unit, configured to perform feature extraction on the effective EEG signal to obtain EEG features;

The prediction unit is used for predicting the EEG feature to obtain a preference prediction result corresponding to the multimedia information.

9. An electronic device, characterized in that the electronic device comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor executes the computer program to cause the electronic device to execute claims 1 to 10. The preference prediction method based on multimedia information according to any one of 7.

10. A readable storage medium, wherein computer program instructions are stored in the readable storage medium, and when the computer program instructions are read and run by a processor, any one of claims 1 to 7 is executed The described preference prediction method based on multimedia information.