CN109815361B - Intelligent music recommendation system based on brain wave identification - Google Patents
Intelligent music recommendation system based on brain wave identification Download PDFInfo
- Publication number
- CN109815361B CN109815361B CN201910088836.4A CN201910088836A CN109815361B CN 109815361 B CN109815361 B CN 109815361B CN 201910088836 A CN201910088836 A CN 201910088836A CN 109815361 B CN109815361 B CN 109815361B
- Authority
- CN
- China
- Prior art keywords
- brain wave
- music
- user
- mobile terminal
- brain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses an intelligent music recommendation system based on brain wave identification, which consists of a wearable Think Gear sensor, a mobile terminal and a background database; the wearable thin Gear sensor is worn on the head of a human body and used for brain wave detection, and the obtained frequency is matched through pattern recognition by frequency analysis of brain wave signals to obtain the mood state of a user; the frequency detected by the wearable Think Gear sensor is sent to the mobile terminal through the Bluetooth module, and the mobile terminal uploads data to the background database through the 4G module; the background database scores according to the uploaded data, and then selects different music according to different moods and scoring values of the user, and plays the music through the mobile terminal. The problem that an existing music recommendation system cannot be accurately adapted to different users is solved.
Description
Technical Field
The invention belongs to the technical field of music recommendation, and particularly relates to an intelligent music recommendation system based on brain wave identification.
Background
The 21 st century is an information age, and along with the development of information technology and network technology, the information is penetrated into various aspects of daily life of people, and close connection is established with the daily life of people. Today, the pace of life is faster and faster, and the demand for entertainment is also stronger and stronger, so various music recommendation systems are also gradually growing. Since ancient times, music has been accompanied by people's daily lives, ancient panpipe suona has been ancient, guitar and saxophone have been so far, and most of people listen to music on line through mobile phones and music players or directly on computers, and mp3 and mp4 are rarely used. The form of music is also continually innovating and evolving. In the early years, people also listen to music through magnetic tapes or compact discs, but digital music is now worldwide, and most of people listen to music through the internet, in other words, the internet is the most popular music medium of digital music at present, and the convenience, data storage security, sharing property, data capacity and the like of a music website combining information technology and internet technology are obviously superior to those of the traditional magnetic tapes and CDs.
The original music recommendation system provides corresponding music according to user preference, but is too extensive, and expected effects may not be achieved. There is a need for a more intelligent system that coordinates different users based thereon, providing the users with more resonated music based on the users' music preferences, search requirements at ordinary times, emotional feedback information, and other certain features.
Disclosure of Invention
The invention aims to overcome the defects of the traditional music recommendation system and provide a more intelligent and convenient intelligent music recommendation system based on brain wave identification for users.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
an intelligent music recommendation system based on brain wave recognition comprises a wearable Think Gear sensor, a mobile terminal and background equipment; the wearable thin Gear sensor is worn on the head of a human body and used for detecting brain wave (eeg) signals, the frequency detected by the wearable thin Gear sensor is sent to a mobile terminal (such as a mobile phone and a tablet personal computer) through a Bluetooth module, and the mobile terminal uploads data to background equipment through a 4G module; the background equipment analyzes and scores according to the brain waves and mood calculation method, and then selects proper music according to analysis results and score values and plays the music through the mobile terminal.
In order to optimize the technical scheme, the specific measures adopted further comprise:
the thin Gear sensor comprises a forehead sensor taking the point position of the earlobe as a reference point, a thin Gear AM chip internally provided with an acquisition, inspection and correction algorithm and a high-sensitivity amplifier, wherein the thin Gear sensor is used for measuring brain wave signals in a digital mode, transmitting brain wave data through a Bluetooth serial port, and carrying out pairing connection through setting a secret key and the same baud rate as Bluetooth, so that transparent transmission of the brain wave serial port data is realized, and wireless transmission signals are output to background equipment.
The background equipment comprises a brain wave acquisition and analysis module, a scoring recommendation module and a music library, wherein the brain wave acquisition and analysis module acquires and analyzes brain wave data, classifies the brain wave data, and the scoring recommendation module plays recommended music and receives user scores according to an analysis result music library.
The brain wave acquisition and analysis module divides brain wave data into four types: delta is 1 ～ 3Hz, θ is 4-7 Hz, α is 8-13 Hz, β is 14-30 Hz, respectively representing sleep state, anxiety state, calm state and excited state.
The scoring algorithm of the scoring recommendation module is as follows:
wherein R is ui A predictive score for user u for music item i;the scoring average value of the user u on all music of the system is obtained; m is M u Scoring the deviation of user u for a certain music item i; t (T) u Scoring the music item i according to brain waves of a user u; k is the specific gravity of the brain wave score in the whole algorithm formula.
M as described above u The calculation formula is as follows:
in sigma u Sum sigma v Deviation values respectively representing scores of target user u and neighbor user v, N i(u) R is a set of neighbor users similar to user u vi Representing the score of user v for item i,and->Representing the average of scores for user u and user v, sim (u, v) is the calculation of pearson correlation coefficient to represent the similarity between users.
T as described above u Calculation ofThe formula is as follows:
in the formula, |x u -λ j | min The minimum positive value of the difference between the brain wave signal value received by the user u and the music threshold value is represented. Wherein x is u Representing the collected user u brain wave number value x u ＝(1,30)Hz；λ j Representing a music threshold, a corresponding EEG signal wave value for music in a corresponding emotional state, where lambda j = {2,5.5,10.5,22}, j takes the values 1,2,3,4. The research shows that brain wave signals generated by human body under normal condition are in the range of 1-30 Hz, different frequencies correspond to different emotions, and the formula is to convert the emotions into analyzable numerical values.
The invention has the following beneficial effects:
after the design of the invention is adopted, the intelligent recommendation of the music is realized to a great extent by detecting the brain waves of the user and timely and accurately recommending the music suitable for the current emotion.
Drawings
FIG. 1 is a software workflow diagram of the present invention;
FIG. 2 is a block diagram of a system of the present invention;
FIG. 3 is a system workflow diagram of the present invention;
FIG. 4 is a block diagram of a brain wave acquisition and transmission flow according to an embodiment of the present invention;
fig. 5 is a graph of brain wave band classification according to the present invention.
Detailed Description
Embodiments of the present invention are described in further detail below with reference to the accompanying drawings.
Referring to fig. 1-3, the intelligent music recommendation system based on brain wave identification comprises a wearable Think Gear sensor, a mobile terminal and background equipment; the wearable Think Gear sensor is worn on the head of a human body for brain wave detection, the frequency detected by the wearable Think Gear sensor is sent to a mobile terminal (such as a mobile phone and a tablet personal computer) through a Bluetooth module, and the mobile terminal uploads data to a background device through a 4G module; the background equipment analyzes and scores according to the brain waves and mood calculation method, and then selects different music according to analysis results and score values to play through the mobile terminal.
In an embodiment, the Think Gear sensor comprises a forehead sensor taking the point position of the earlobe as a reference point, a Think Gear AM chip internally provided with an acquisition, inspection and correction algorithm and a high-sensitivity amplifier, brain wave signals measured by the Think Gear sensor in a digital mode are transmitted through a Bluetooth serial port, and the Think Gear sensor is connected in a pairing way through setting a secret key and a baud rate which are the same as those of Bluetooth, so that transparent transmission of the brain wave serial port data is realized, and wireless transmission signals are output to background equipment.
In the brain wave acquisition and transmission process, referring to fig. 4, the wearable device Think Gear brain wave sensor is used for acquiring brain wave signals and transmitting the acquired brain wave signals to the background device. The brain wave signals are converted into electric signals, then the electric signals are transmitted to the background through Bluetooth to finish the processing of the signals, and the signals are output to the mobile terminal for matching. The general process of converting EEG signals into electrical signals is: outputting serial data to CPLD (programmable logic device) for temporary storage; the CPLD converts serial data into parallel data and sends the parallel data to an MCU (micro controller) through a multiplexer; finally, the MCU (microcontroller) transmits the data to the Bluetooth module, through which the data is wirelessly transmitted.
The control logic in the CPLD (programmable logic device) can determine whether the ARM continues to read the current data according to the transmitted data and the state of the ARM reading the data. Meanwhile, in order to facilitate data processing, the number of CPLD and ARM interfaces is reduced, serial data output by each ADC is firstly converted into parallel data, then the data is output to the ARM through a multiplexer, and finally the data is transmitted to the Bluetooth module through a UARTI port.
After the processing of the signal is completed, wireless bluetooth transmission is performed next. The Bluetooth transmission, namely the Bluetooth module is only communicated with the ARM, and the ARM transmits data to the Bluetooth module through the UART1 interface and then sends the data wirelessly. The UART1 interface is an interface provided by a Bluetooth embedded module fast BTM0704C2P chip. The Bluetooth module communicates with the LPC2144 through the UART, and the Bluetooth module wirelessly transmits the received data to the embedded processing terminal. Wherein LPC2144 is the master control chip used as CPLD.
In an embodiment, the background device comprises a brain wave acquisition and analysis module, a scoring recommendation module and a music library, wherein the brain wave acquisition and analysis module acquires and analyzes brain wave data, classifies the brain wave data, and the scoring recommendation module recommends music to play music and receives a user score according to an analysis result.
The research shows that brain waves are a method for recording brain activities by using an electrophysiological index, and postsynaptic potentials generated by a large number of neurons synchronously occur after the summation of the postsynaptic potentials during the brain activities. It records the changes of electric wave during brain activity, which are spontaneous rhythmic nerve electric activity, and its frequency range is usually between 1-30 times per second, and it can be divided into four wave bands, namely delta wave, theta wave, alpha wave and beta wave. Delta wave: the frequency is 1-3 Hz and the amplitude is 20-200 mu V. Such bands can be recorded on the temporal and parietal lobes when the person is in infancy or mental retardation, and the adult is in extreme fatigue and comatose or anesthetized state; theta wave: the frequency is 4-7 Hz and the amplitude is 5-20 mu V. Such waves are extremely pronounced in adult patients who are intended to be frustrated or depressed. But this wave is the main component in the electroencephalogram of teenagers (10-17 years old); alpha wave: the frequency is 8-13 Hz (average is 10 Hz), and the amplitude is 20-100 mu V. It is the fundamental rhythm of normal human brain waves, whose frequency is fairly constant if no stimulus is applied. The rhythm is most obvious when the person wakes up, is quiet and closes eyes, because the waveform is closest to the brain electrical biological rhythm of the right brain, and the inspiration state of the person appears; beta wave: the frequency is 14-30 Hz, and the amplitude is 100-150 mu V. This wave occurs when the stress and emotion are excited or stimulated, and when the person wakes up from nightmares, the original slow wave rhythm can be replaced immediately by the rhythm. In addition, brain waves with special waveforms such as hump waves, sigma waves, lambda waves, kappa-compound waves, mu waves and the like can appear during sleeping.
Therefore, in the embodiment of the invention, the brain wave acquisition and analysis module classifies brain wave data into four types: delta is 1-3 Hz, theta is 4-7 Hz, alpha is 8-13 Hz, beta is 14-30 Hz, and the sleeping state, the worry state, the calm state and the excited state of a natural person are respectively shown, and the division is shown in figure 5.
In an embodiment, the scoring algorithm of the scoring recommendation module is:
wherein R is ui The predictive score of the user u for the music item i is represented, and the higher the score, the more preferentially the music item is recommended.Representing the average value of scores of user u on all music of the system; m is M u A bias score representing the user u for a certain music item i; t (T) u Representing the score of the music item i obtained according to the brain waves of the user u, selecting lambda value and lambda based on the corresponding emotion by combining the analysis result of the EEG signals obtained by the brain wave acquisition and analysis module j Representing a music threshold (lambda) j = {2,5.5,10.5,22}, j takes the value 1,2,3, 4), and the music threshold represents the music type corresponding to the corresponding emotional state at a certain EEG signal wave value.
For example: lambda (lambda) 1 When the music threshold value is 2, selecting music suitable for sleep state; when lambda is 2 ＝5.5，λ 3 ＝10.5，λ 4 When the value of the music threshold is 5.5,10.5 and 22, selecting music suitable for the state of anxiety, calm or excited state; when the brain wave signal result of the background analysis is in a sleep state, lambda is taken out 1 A value of 2; when the brain wave signal result of background analysis is in a state of anxiety, calm or excited state, lambda is taken respectively j Values 5.5,10.5, 22; then combining the brain wave signal value with the music threshold value to calculate the brain wave value (x u ) And a music threshold lambda j The absolute value of the difference is minimized, the smaller the difference is, which means that the current emotion state of the user is closer to the music threshold, the higher the music accuracy that is recommended to be suitable for the current emotion of the user is, because of the T calculated in this case u Larger. k is used for adjusting the proportion of brain wave scores in the whole algorithm formulaThe larger k represents the larger the brain wave score ratio, the more prominent the brain wave recommendation advantages, the final score is calculated by integrating the whole algorithm, and the background system selects high-score music to be recommended to the user preferentially.
In an embodiment, M u The calculation formula is as follows:
in sigma u Sum sigma v Deviation values respectively representing scores of target user u and neighbor user v, N i(u) R is a set of neighbor users similar to user u vi Representing the score of user v for item i,and->Representing the average of scores for user u and user v, sim (u, v) is the calculation of pearson correlation coefficient to represent the similarity between users.
With respect to the pearson correlation coefficient,wherein r is ui 、r vi Actual scoring of item i for users u and v, respectively,/->And->Mean value of scores representing user u and user v, I u And I v Scoring sets of items for user u and user v, respectively, using a common score between users to measure their similarity, each user scored with a vector R n×m The numbers of n and m are the number of users and the number of items, respectively. The similarity between users can be calculated by calculating the distance between the vectors.
In the examples, T u The calculation formula is as follows:
in the formula, |x u -λ j | min The minimum positive value of the difference between the brain wave signal value received by the user u and the music threshold value is represented. Wherein x is u Representing the collected user u brain wave number value x u ＝(1,30)Hz；λ j Representing a music threshold, a corresponding EEG signal wave value for music in a corresponding emotional state, where lambda j = {2,5.5,10.5,22}. The brain wave signals generated by the human body under normal conditions are in the range of 1-30 hz, different frequencies correspond to different emotions, and the scoring algorithm of the scoring recommendation module is used for converting the emotions into scores. The comprehensive score of a user on music library is obtained through calculation of the algorithm formula, the action of the brain waves in the recommendation system can be adjusted according to the proportion of the brain wave score, so that information matching is more accurate, all scores are synthesized, and priority recommendation is performed according to the measured score. The higher the score, the better the recommendation.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to the invention without departing from the principles thereof are intended to be within the scope of the invention as set forth in the following claims.
Claims (4)
1. An intelligent music recommendation system based on brain wave recognition is characterized in that: the wearable thin Gear sensor comprises a wearable thin Gear sensor, a mobile terminal and background equipment; the wearable Think Gear sensor is worn on the head of a human body and used for brain wave detection; the electric signal detected by the wearable thin Gear sensor is converted into frequency output through a frequency discrimination link, the frequency is sent to the mobile terminal through a Bluetooth module, and the mobile terminal uploads data to background equipment through a 4G module; the background equipment analyzes and scores according to the brain waves and mood calculation method, then selects different music according to analysis results and score values, and plays the music through the mobile terminal, and specifically:
firstly, brain wave acquisition and analysis are carried out, score calculation is carried out based on emotion types after emotion types are obtained, and finally music selection is realized;
wherein, the score calculation formula is:
wherein R is ui A predictive score for user u for music item i;the scoring average value of the user u on all music of the system is obtained; m is M u Scoring the deviation of user u for a certain music item i; t (T) u Scoring the music item i according to brain waves of a user u; k is the proportion of brain wave scores in the whole algorithm formula;
in sigma u Sum sigma v Deviation values respectively representing scores of target user u and neighbor user v, N i(u) R is a set of neighbor users similar to user u vi Representing the score of user v for item i,representing the average of scores of users v, sim (u, v) is the calculation of pearson correlation coefficients to represent the similarity between users;
in the formula, |x u -λ j | min The minimum positive value of the difference between the brain wave signal value received by the user u and the music threshold value is obtained; x is x u For the collected user u brain wave number value x u ＝(1,30)Hz；λ j Is a music threshold value-different emotion state soundCorresponding brain wave signal wave value, wherein lambda j = {2,5.5,10.5,22}, j takes the values 1,2,3,4.
2. The intelligent music recommendation system based on brain wave recognition according to claim 1, wherein: the thin Gear sensor comprises a forehead sensor taking the point position at the earlobe as a reference point, a thin Gear AM chip internally provided with an acquisition, inspection and correction algorithm and a high-sensitivity amplifier, wherein the thin Gear sensor is used for measuring brain wave signals in a digital mode, transmitting brain wave data through a Bluetooth serial port, and carrying out pairing connection through setting a secret key and a baud rate which are the same as those of Bluetooth, so that transparent transmission of the brain wave serial port data is realized, and wireless transmission signals are output to background equipment.
3. The intelligent music recommendation system based on brain wave recognition according to claim 2, wherein: the background equipment comprises a brain wave acquisition and analysis module, a scoring recommendation module and a music library, wherein the brain wave acquisition and analysis module acquires and analyzes brain wave data, classifies the brain wave data, and the scoring recommendation module plays recommended music according to an analysis result music library and receives user scores.
4. The intelligent music recommendation system based on brain wave recognition according to claim 3, wherein: the brain wave acquisition and analysis module divides brain wave data into four types: delta is 1-3 Hz, theta is 4-7 Hz, alpha is 8-13 Hz, beta is 14-30 Hz, and the sleeping state, the anxiety state, the calm state and the excited state are respectively represented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910088836.4A CN109815361B (en) | 2019-01-29 | 2019-01-29 | Intelligent music recommendation system based on brain wave identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910088836.4A CN109815361B (en) | 2019-01-29 | 2019-01-29 | Intelligent music recommendation system based on brain wave identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109815361A CN109815361A (en) | 2019-05-28 |
| CN109815361B true CN109815361B (en) | 2023-08-08 |
Family
ID=66605723
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910088836.4A Active CN109815361B (en) | 2019-01-29 | 2019-01-29 | Intelligent music recommendation system based on brain wave identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109815361B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110801568B (en) * | 2019-06-28 | 2022-07-08 | 林万佳 | Device for making and outputting pulsating music and method for forming pulsating music |
| CN113220122A (en) * | 2021-05-06 | 2021-08-06 | 西安慧脑智能科技有限公司 | Brain wave audio processing method, equipment and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103412646A (en) * | 2013-08-07 | 2013-11-27 | 南京师范大学 | Emotional music recommendation method based on brain-computer interaction |
| CN106991122A (en) * | 2017-02-27 | 2017-07-28 | 四川大学 | A kind of film based on particle cluster algorithm recommends method |
| CN108073284A (en) * | 2017-12-15 | 2018-05-25 | 南京信息工程大学 | Purchase system based on brain wave identification mood |
-
2019
- 2019-01-29 CN CN201910088836.4A patent/CN109815361B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103412646A (en) * | 2013-08-07 | 2013-11-27 | 南京师范大学 | Emotional music recommendation method based on brain-computer interaction |
| CN106991122A (en) * | 2017-02-27 | 2017-07-28 | 四川大学 | A kind of film based on particle cluster algorithm recommends method |
| CN108073284A (en) * | 2017-12-15 | 2018-05-25 | 南京信息工程大学 | Purchase system based on brain wave identification mood |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109815361A (en) | 2019-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bi et al. | AutoDietary: A wearable acoustic sensor system for food intake recognition in daily life | |
| Yu et al. | WiFi-sleep: Sleep stage monitoring using commodity Wi-Fi devices | |
| Mendonca et al. | A review of obstructive sleep apnea detection approaches | |
| Zhang et al. | Monitoring chewing and eating in free-living using smart eyeglasses | |
| WO2020119245A1 (en) | Wearable bracelet-based emotion recognition system and method | |
| Ren et al. | Fine-grained sleep monitoring: Hearing your breathing with smartphones | |
| Haidar et al. | Convolutional neural networks on multiple respiratory channels to detect hypopnea and obstructive apnea events | |
| US20160302671A1 (en) | Prediction of Health Status from Physiological Data | |
| WO2018103028A1 (en) | Audio playback device, system, and method | |
| CN105807674A (en) | Intelligent wearable device capable of controlling audio frequency terminal and control method | |
| Dai et al. | Wearable biosensor network enabled multimodal daily-life emotion recognition employing reputation-driven imbalanced fuzzy classification | |
| US20140330132A1 (en) | Physiological characteristic detection based on reflected components of light | |
| WO2017069644A2 (en) | Wireless eeg headphones for cognitive tracking and neurofeedback | |
| AU2013256179A1 (en) | Physiological characteristic detection based on reflected components of light | |
| KR102015097B1 (en) | Apparatus and computer readable recorder medium stored program for recognizing emotion using biometric data | |
| CN109815361B (en) | Intelligent music recommendation system based on brain wave identification | |
| Niu et al. | Research on genetic algorithm based on emotion recognition using physiological signals | |
| CN110301890B (en) | Apnea monitoring method and device | |
| Yan et al. | Emotion classification with multichannel physiological signals using hybrid feature and adaptive decision fusion | |
| CN106562762A (en) | Sleep monitoring and improving apparatus and method | |
| CN107239769A (en) | A kind of personal emotion potency recognition methods of use multi-channel information synchronization | |
| Tran et al. | Stethoscope-sensed speech and breath-sounds for person identification with sparse training data | |
| Cai et al. | Application of electroencephalography-based machine learning in emotion recognition: A review | |
| Camcı et al. | Abnormal respiratory event detection in sleep: A prescreening system with smart wearables | |
| Shen et al. | Multitask residual shrinkage convolutional neural network for sleep apnea detection based on wearable bracelet photoplethysmography |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |