CN106611028A - Personalized music recommendation method and system - Google Patents
Personalized music recommendation method and system Download PDFInfo
- Publication number
- CN106611028A CN106611028A CN201510708821.5A CN201510708821A CN106611028A CN 106611028 A CN106611028 A CN 106611028A CN 201510708821 A CN201510708821 A CN 201510708821A CN 106611028 A CN106611028 A CN 106611028A
- Authority
- CN
- China
- Prior art keywords
- music
- user
- xth
- physiological data
- song
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/63—Querying
- G06F16/635—Filtering based on additional data, e.g. user or group profiles
- G06F16/636—Filtering based on additional data, e.g. user or group profiles by using biological or physiological data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Physiology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a personalized music recommendation method comprising the following steps of acquiring various physiological data of a user before listening to music; acquiring the various physiological data of the user after listening to music; acquiring a negative label, a positive label or a seemingly positive label of each physiological data of each song by listening to music; comparing the various physiological data of the user before and after listening to music, and analyzing physiological and emotional response degrees of the user to the music; and recommending the music to the user in a personalized manner according to the physiological and emotional response degrees of the user to different music. According to the method provided by the invention, how the user likes the music is surveyed by directly using the physiological data when the user listens to the music, subjective feeling of the user can be acquired in an objective manner, and a brand new personalized music recommendation path is provided for a music industry; and meanwhile, the physiological data, such as a heart rate, a respiratory rate and a skin electric conductance are adopted as research objects, so that surveying on the emotion of the user is more reliable.
Description
Technical field
The present invention relates to intelligence wearing, music industry, musical therapy technology, more particularly to a kind of music individual character
The recommendation method and system of change.
Background technology
With the development of digital music, traditional CD player, walkman is substituted by mobile phone and computer.Respectively
On the one hand class music platform and music software provide convenience for the Music Appreciation of user, on the other hand also cause
The problem of music personalized recommendation is increasingly highlighted：In the face of hundreds of thousands easily, in addition music up to a million, user
It is difficult to the music oneself liked therefrom is retrieved, and music software side is same facing to how to user individual
Recommend a difficult problem for music.So, the personalized recommendation of music becomes more and more important, is not only related to user's
Music Appreciation is experienced, and concerns the survival and development of music software side.
Traditional music personalized recommendation method typically passes through the music preferences of questionnaire survey user, by investigating
Broadcasting integrity, broadcasting time of the user to a piece of music, and by music-related information (such as musical instrument class
Type, music style, composer, musician etc.) recommended.But above-mentioned music personalized recommendation method
Process is complicated, and is difficult to find the music that user really likes accurately.
Research shows that music can affect the emotion of people, and then affect the physiology of people.So, when user is glad
During the music that reward is liked oneself, subjective emotion changes will be emerged from by objective physiological change.Root
According to the research over 100 years of music cognition field nearly, when people enjoy the music oneself liked, can produce
Significant physiologic response, be mainly reflected in heart rate, respiratory frequency, skin conductivity, body temperature, blood pressure, blood oxygen,
In the physical signs such as muscle tone, biomolecule.It is, in general, that the music of incentive can increase user's
Heart rate, respiratory frequency and skin conductivity；And sedative music can reduce the heart rate of user, respiratory frequency
And skin conductivity.However, above-mentioned basic research is merely resting on the experimental stage, not yet start practical application,
More it is applied not on the personalized recommendation of music.
Chinese Patent Application No. is：201410338956.2, it is entitled：A kind of music recommends method and system life
Thing patent of invention, the emotion of user is analyzed based on physiological health data, so as to recommend identical emotion for user
The music of label.Its method is as follows：The history health data of user is wanted to compare with current health data,
The emotion label of active user is calculated, then the music with identical emotion label in music libraries is recommended into use
Family.The core methed of the patent is the emotion label of music in the emotion label and music libraries for match user, will
The active of user is submitted to, becomes automatic data collection and analysis.This method is compared other traditional music and is recommended
Method has certain progressive, but still falls within a kind of labelled mode, it is difficult to ensure that the music recommended
It is exactly music that user likes.Such as, when user is all in getting excited or being depressed, it is impossible to all can
Like the music of all respective labels in music libraries.So above-mentioned patent can not be investigated directly with physiological data
Fancy grade of the user to music.And the physiology that the Study on Physiologic and music of above-mentioned patent investigation affects refers to
Mark is less related.The physical signs of the patent are family electrocardiogram (ECG) data, user's blood pressure data, user's temperature data.
However, according to the basic research over nearly 100 years, music mainly affects heart rate, breathing, the skin conductivity of human body
Deng physiological data, so above-mentioned patent can not well investigate impact of the music to human body, use can not be investigated
Fancy grade of the family to music.
The content of the invention
This patent provides a kind of personalized recommendation method of music of the Music Appreciation experience for lifting user and is
System.
For achieving the above object, the present invention adopts following technical proposals：
A kind of personalized recommendation method of music, comprises the steps：
Obtain user and listen to the every physiological data before music；
Obtain every physiological data during user's listening music；
By listening to music, obtain " feminine gender ", " positive " of items physiological data or " doubtful per a piece of music
Like positive " label；
Contrast user listens to the every physiological data before and after music, and analyzing user should to the physiology and emotion of music
Answer degree；
According to physiology and emotion response degree of the user to different music, to user individualized music recommendation is carried out.
On the other hand, wherein, " obtain user listen to music before physiological data ", comprise the steps：
With the bioimpedance data that bio-impedance sensor obtains user；
The physiological datas such as heart rate, respiratory frequency, the skin conductivity of user are calculated with the bioimpedance data for obtaining；
Before user listens to music, on an hourly basis the frequency of 6 times, gathers the physiology number under user's quiescent condition
According to：Heart rate samples, respiratory frequency sample and skin conductivity sample, achieve stand-by.
On the other hand, wherein, " obtain user listen to music when physiological data ", comprise the steps：
With the bioimpedance data that bio-impedance sensor obtains user；
The physiological datas such as heart rate, respiratory frequency, the skin conductivity of user are calculated with the bioimpedance data for obtaining；
When user listens to music, by the frequency collection user of 6 times per minute xth (x is >=1 integer) is listened
Action physiological data during song：Heart rate samples, respiratory frequency sample and skin conductivity sample, prepare into
Row next step data analysiss.
In certain embodiments, the bio-impedance sensor can also be replaced using optoelectronic device.
In certain embodiments, the physiological data includes but is not limited to heart rate, breathing, the skin pricktest of user
Lead, the data such as body temperature, blood pressure, blood oxygen, muscle tone or biomolecule.
In certain embodiments, the score value of the physiological data include but is not limited to 100 points system, ten point system or
1 point of system.
On the other hand, wherein, " contrast user listen to music before and after every physiological data, analyze user couple
The physiology and emotion response degree of music ", comprises the steps：
All music all obtain 40 points of basis point；
Skin conductivity score：If xth song is " sun in the label of " skin conductivity " this physiological data
Property " (25-100PSx) point is then obtained, it is that " the doubtful positive " then obtains (25-100PSx)/2 point, it is " cloudy
Property " then obtain 0 point；
Respiratory frequency score：If xth song is " sun in the label of " respiratory frequency " this physiological data
Property " (20-100PRx) point is then obtained, it is that " the doubtful positive " then obtains (20-100PRx)/2 point, it is " cloudy
Property " then obtain 0 point；
Heart rate score：Obtain if xth song is " positive " in the label of " heart rate " this physiological data
(15-100PHx) point is obtained, is that " the doubtful positive " then obtains (15-100PHx)/2 point, be " feminine gender "
Then obtain 0 point；
By the public affairs on " total score=basis point+skin conductivity score+respiratory frequency score+heart rate score "
Formula calculates the first music total score of xth.
On the other hand, wherein, " according to physiology and emotion response degree of the user to different music, to user
Carry out individualized music recommendation ", comprise the steps：
User's xth (x >=2) is secondary appreciate music when, with (x-1) secondary music total score by arranging from high to low
Row；
The higher music of fraction, recommended frequency is higher, and fraction is lower, and recommended frequency is lower, recommends every time
The new music that a certain proportion of user never listened, music arrangement mode is random；
While user's xth (x >=2) secondary appreciation music, carry out the collection of the physiological data of a new round and comment
Point；
By the music total score that the scoring of all music of the secondary appreciation of user's xth (x >=2) is secondary with former (x-1)
Integrated, to live the music listened, sorting again after averaging, to the music do not listened in the past, directly
Connect the fraction obtained with this and participate in sequence；
Xth (x >=2) wheel sequence is carried out to the music that user appreciates by score from big to small；
Using xth wheel music ranking results as the personalized recommendation foundation of ((x+1)th time) next time.
A kind of personalized recommendation of music, including：
Physiological data collection module, for when user does not listen to music or listens to music, gathering the life of user
Reason data, while setting up " tranquillization physiological data storehouse " and " action physiological data storehouse "；
Musical taste grading module, enters respectively to " tranquillization physiological data storehouse " and " action physiological data storehouse "
Row analysis, obtains " the individualized music data base " of user；
Individualized music recommending module, individualized music is carried out with reference to " individualized music data base " to user
Recommend.
Beneficial effects of the present invention are：The present invention is directly investigated with physiological data when user's appreciation music
Favorable rating of the user to music, can obtain the subjective feeling of user by objective mode, be music row
Industry provides a brand-new music personalized recommendation path；Meanwhile, the present invention choose heart rate, respiratory frequency,
The physiological datas such as skin conductivity are used as object of study, and the investigation to user emotion is relatively reliable.
Description of the drawings
Fig. 1 is the overall flow figure of the recommendation method of the music personalization that the present invention is provided.
Fig. 2 is the flow chart for obtaining the physiological data before user's listening music that the present invention is provided.
The flow chart of physiological data when Fig. 3 is acquisition user's listening music of present invention offer.
Fig. 4 is that the per song of the recommendation method of the music personalization that the present invention is provided affects user's physiological parameter
Score calculate figure.
Fig. 5 be the present invention provide carry out to user scheme the step of individualized music is recommended.
Fig. 6 is the construct pattern figure of the commending system of the music personalization that the present invention is provided.
Specific embodiment
In order that the purpose of the present invention, technical scheme and beneficial effect become more apparent, below in conjunction with accompanying drawing
And embodiment, the present invention will be described in further detail.It should be appreciated that described herein be embodied as
Example is not intended to limit the present invention only to explain the present invention.
As shown in figure 1, a kind of personalized recommendation method of music, comprises the steps：
Obtain user and listen to the every physiological data before music；
Obtain every physiological data during user's listening music；
By listening to music, obtain " feminine gender ", " positive " of items physiological data or " doubtful per a piece of music
Like positive " label；
Contrast user listens to the every physiological data before and after music, and analyzing user should to the physiology and emotion of music
Answer degree；
According to physiology and emotion response degree of the user to different music, to user individualized music recommendation is carried out.
Referring again to Fig. 2, wherein, " obtain user and listen to the physiological data before music " comprises the steps：
With the bioimpedance data that bio-impedance sensor obtains user；
The physiological datas such as heart rate, respiratory frequency, the skin conductivity of user are calculated with the bioimpedance data for obtaining；
Before user listens to music, on an hourly basis the frequency of 6 times, gathers the physiology number under user's quiescent condition
According to：Heart rate samples, respiratory frequency sample and skin conductivity sample, achieve stand-by.
Referring again to Fig. 3, wherein, " obtaining physiological data during user's listening music " comprises the steps：
With the bioimpedance data that bio-impedance sensor obtains user；
The physiological datas such as heart rate, respiratory frequency, the skin conductivity of user are calculated with the bioimpedance data for obtaining；
When user listens to music, by the frequency collection user of 6 times per minute xth (x is >=1 integer) is listened
Action physiological data during song：Heart rate samples, respiratory frequency sample and skin conductivity sample, prepare into
Row next step data analysiss.
In certain embodiments, physiological data collection mode of the invention with bio-impedance device except being obtained
Outside the physiological datas such as heart rate, respiratory frequency, skin conductivity, related data can also be in other ways obtained,
Such as heart rate is obtained with optoelectronic device, during practical operation, can unrestricted choice as needed.
In certain embodiments, physiological data include but is not limited to the heart rate of user, breathing, skin conductivity,
The data such as body temperature, blood pressure, blood oxygen, muscle tone or biomolecule.
In certain embodiments, the score value of physiological data includes but is not limited to 100 points of systems, ten point systems or 1
Divide system, meanwhile, in the present invention, the best result ratio shared by each physiological data can be 1：1：1,
Can be 3：2：1 or 2：2：1.
Specific to, impact of the xth song to user is analyzed with " skin conductivity ", comprise the steps：
S1：User is listened skin conductivity sample SSx during xth song, with skin conductivity sample during tranquillization
This SS0 does T inspections.T is checked, and is mainly used in less (such as x of sample content<30), population standard deviation
σ unknown normal distribution data.T inspection be the probability occurred come inference difference with t-distribution theory, so as to than
Whether notable compared with the difference of two averages, this sentences T and checks the P values for obtaining as music to user emotion
The basis for estimation of influence degree.
S2：If the user while when listening xth song, skin conductivity parameter is joined with skin conductivity under quiescent condition
P value PSx >=0.05 that number T inspections are obtained, represents that the song is not produced significantly to the skin conductivity of user
Affect, user may not like this song, therefore the impact by the song to user's skin conductivity is judged as
It is negative；
S3：If 0.01 ＜ PSx ＜ 0.05, then it represents that the song generates aobvious to the skin conductivity of user
Writing affects, and user may prefer this song；
S4：If PSx≤0.01, represent that the song generates pole appreciable impact to the skin conductivity of user,
User may enjoy a lot this song；
S5：By the music collection of all PSx ＜ 0.05, next step analysis is ready for；
S6：The PSx of xth song in selecting above-mentioned S5 to gather successively, and the song (xth of xth -1
The previous head of song) PS (x-1), prepare to compare；
S7：If the PS (x-1) >=0.05 in above-mentioned S6, illustrate the song of xth -1 that user listened not
Appreciable impact is generated on the skin conductivity of user, illustrates that user's skin conductivity when xth song is listened is subject to
Appreciable impact really by caused by xth song, rather than affected by the song of xth -1, then by
Impact of the x songs to user's skin conductivity is judged as the positive；
S8：If PS (x-1) ＜ 0.05 in above-mentioned S6, and the action skin that PS (x-1) and PSx is located
Conductance deviates from different directions (two kinds of situations of tranquillization skin conductivity：Compared with quiescent condition, xth head sounds
Pleasure increased user's skin conductivity, and the song of xth -1 reduces the skin conductivity of user；Or xth is first
Music reduces user's skin conductivity, and the song of xth -1 increased the skin conductivity of user), illustrate to use
The family appreciable impact that skin conductivity is subject to when xth song is listened really by caused by xth song, and not
It is to be affected by the song of xth -1, the then impact by xth song to user's skin conductivity is judged as the positive；
S9：If PS (x-1) ＜ 0.05 in above-mentioned S6, and the skin conductivity that PS (x-1) and PSx is located
Deviate from equidirectional (two kinds of situations of skin conductivity under quiescent condition：Compared with quiescent condition, xth is first
Music and the song of xth -1 both increase user's skin conductivity, or the skin conductivity for all reducing user),
And PS (x-1) ＞ PSx, show impact of the xth song to user's skin conductivity, than the song of xth -1
Impact to user's skin conductivity is stronger, and the then impact by xth song to user's skin conductivity judges
For the positive；
S10：If PS (x-1) ＜ 0.05 in above-mentioned S6, and the skin pricktest that PS (x-1) and PSx is located
Lead equidirectional (the two kinds of situations for deviating from skin conductivity under quiescent condition：Compared with quiescent condition, xth
Song and the song of xth -1 both increase user's skin conductivity, or the skin conductivity for all reducing user),
And PS (x-1)≤PSx, illustrate appreciable impact of the xth song on user's skin conductivity, it may be possible to due to
The false positive that the song of xth -1 is caused on the Carry-over effect that user's skin conductivity affects strongly, so by xth
Impact of the song to user's skin conductivity is judged as the doubtful positive；
So far, this every a piece of music for being appreciated of user is all marked with skin conductivity this physiological data
" feminine gender ", " positive " or " the doubtful positive " label, proceed the analysis to " respiratory frequency ".
Specific to, impact of the xth song to user is analyzed with " respiratory frequency ", comprise the steps：
S11：User is listened respiratory frequency sample SRx during xth song, with respiratory frequency sample during tranquillization
SR0 does T inspections.T checks (Student's t test), is mainly used in sample content less (such as n ＜ 30),
Population standard deviation σ unknown normal distribution data.T inspections are the probability occurred come inference difference with t-distribution theory,
So as to whether the difference for comparing two averages is notable, this sentences T and checks the P values for obtaining as music to user
The basis for estimation of emotion influence degree.
S12：If the user while when listening xth song, respiratory frequency parameter is joined with respiratory frequency under quiescent condition
P value PRx >=0.05 that number T inspections are obtained, represents that the song does not make a significant impact on the respiratory frequency of user,
User may not like this song, therefore the impact by the song to user's respiratory frequency is judged as feminine gender；
S13：If 0.01 ＜ PRx ＜ 0.05, then it represents that the song is generated significantly to the respiratory frequency of user
Affect, user may prefer this song；
S14：If PRx≤0.01, represent that the song generates pole appreciable impact to the respiratory frequency of user,
User may enjoy a lot this song；
S15：By the music collection of all PRx ＜ 0.05, next step analysis is ready for；
S16：The PRx of xth song in selecting above-mentioned S15 to gather successively, and the song of xth -1 (xth head
The previous head of music) PR (x-1), prepare to compare；
S17：If the PR (x-1) >=0.05 in above-mentioned S16, illustrate that the song of xth -1 that user is listened is not right
The respiratory frequency of user generates appreciable impact, illustrates that user's respiratory frequency when xth song is listened is subject to aobvious
Writing affects really by caused by xth song, rather than is affected by the song of xth -1, then by xth song
Impact to user's respiratory frequency is judged as the positive；
S18：If PR (x-1) ＜ 0.05 in above-mentioned S16, and the action breathing that PR (x-1) and PRx is located
Frequency departure is in different directions (two kinds of situations of tranquillization respiratory frequency：Compared with quiescent condition, xth song
User's respiratory frequency is increased, and the song of xth -1 reduces the respiratory frequency of user；Or xth song drop
Low user's respiratory frequency, and the song of xth -1 increased the respiratory frequency of user), illustrate that user is listening xth
The appreciable impact that respiratory frequency is subject to during song really by caused by xth song, rather than by the first sound of xth -1
Happy to affect, the then impact by xth song to user's respiratory frequency is judged as the positive；
S19：If PR (x-1) ＜ 0.05 in above-mentioned S16, and the respiratory frequency that PR (x-1) and PRx is located
Deviate from equidirectional (two kinds of situations of respiratory frequency under quiescent condition：Compared with quiescent condition, xth head sounds
The happy song of xth -1 both increases user's respiratory frequency, or the respiratory frequency for all reducing user), and
PR (x-1) ＞ PRx, show impact of the xth song to user's respiratory frequency, and user is exhaled than the song of xth -1
The impact for inhaling frequency is stronger, and the then impact by xth song to user's respiratory frequency is judged as the positive；
S20：If PR (x-1) ＜ 0.05 in above-mentioned S16, and the respiratory frequency that PR (x-1) and PRx is located
Deviate from equidirectional (two kinds of situations of respiratory frequency under quiescent condition：Compared with quiescent condition, xth head sounds
The happy song of xth -1 both increases user's respiratory frequency, or the respiratory frequency for all reducing user), and
PR (x-1)≤PRx, illustrates appreciable impact of the xth song on user's respiratory frequency, it may be possible to due to xth -1
The false positive that song is caused on the Carry-over effect that user's respiratory frequency affects strongly, so by xth song pair
The impact of user's respiratory frequency is judged as the doubtful positive；
So far, this every a piece of music for being appreciated of user is all marked with respiratory frequency this physiological data
" feminine gender ", " positive " or " the doubtful positive " label, proceed the analysis to " heart rate ".
Specific to, impact of the xth song to user is analyzed with " heart rate ", comprise the steps：
S21：User is listened heart rate samples SHx during xth song, with the heart rate samples SH0 during tranquillization T is
Inspection.T checks (Student's t test), is mainly used in sample content less (such as n ＜ 30), overall standard
Difference σ unknown normal distribution data.T inspection be the probability occurred come inference difference with t-distribution theory, so as to than
Whether notable compared with the difference of two averages, this sentences T and checks the P values for obtaining as music to user emotion shadow
The basis for estimation of the degree of sound.
S22：If the user while when listening xth song, hrv parameter is obtained with hrv parameter T inspections under quiescent condition
P value PHx >=0.05 for obtaining, represents that the song does not make a significant impact on the heart rate of user, and user may not
Like this song, therefore the impact by the song to user's heart rate is judged as feminine gender；
S23：If 0.01 ＜ PHx ＜ 0.05, then it represents that the song generates appreciable impact to the heart rate of user,
User may prefer this song；
S24：If PHx≤0.01, represent that the song generates pole appreciable impact on the heart rate of user, use
Family may enjoy a lot this song；
S25：By the music collection of all PHx ＜ 0.05, next step analysis is ready for；
S26：The PHx of xth song in selecting above-mentioned S25 to gather successively, and the song of xth -1 (xth head
The previous head of music) PH (x-1), prepare to compare；
S27：If the PH (x-1) >=0.05 in above-mentioned S26, illustrate that the song of xth -1 that user is listened is not right
The heart rate of user generates appreciable impact, illustrates that user's appreciable impact that heart rate is subject to when xth song is listened is true
In fact by caused by xth song, rather than affected by the song of xth -1, then by xth song to user's heart
The impact of rate is judged as the positive；
S28：If PH (x-1) ＜ 0.05 in above-mentioned S26, and the action heart rate that PH (x-1) and PHx is located
Deviate from different directions (two kinds of situations of resting heart rate：Compared with quiescent condition, xth song increased use
Family heart rate, and the song of xth -1 reduces the heart rate of user；Or xth song reduces user's heart rate, and
The song of xth -1 increased the heart rate of user), illustrate the notable shadow that user's heart rate when xth song is listened is subject to
Ring really by caused by xth song, rather than affected by the song of xth -1, then by xth song to
The impact of family heart rate is judged as the positive；
S29：If PH (x-1) ＜ 0.05 in above-mentioned S26, and the heart rate deviation that PH (x-1) and PHx is located
Equidirectional (two kinds of situations of heart rate under quiescent condition：Compared with quiescent condition, xth song and xth -1
Song both increases user's heart rate, or the heart rate for all reducing user), and PH (x-1) ＞ PHx, table
Impact of the bright xth song to user's heart rate, it is stronger than impact of the song of xth -1 to user's heart rate, then
Impact of the xth song to user's heart rate is judged as into the positive；
S30：If PH (x-1) ＜ 0.05 in above-mentioned S26, and the heart rate deviation that PH (x-1) and PHx is located
Equidirectional (two kinds of situations of heart rate under quiescent condition：Compared with quiescent condition, xth song and xth -1
Song both increases user's heart rate, or the heart rate for all reducing user), and PH (x-1)≤PHx, say
Appreciable impact of the bright xth song on user's heart rate, it may be possible to because the song of xth -1 is to the strong shadow of user's heart rate
The false positive that loud Carry-over effect is caused, so the impact by xth song to user's heart rate is judged as doubtful sun
Property；
So far, this every a piece of music for being appreciated of user is all marked with " cloudy in heart rate this physiological parameter
Property ", " positive " or " the doubtful positive " label.
Referring again to Fig. 4, on this basis, " contrast user listens to the every physiological data before and after music, divides
Physiology and emotion response degree of the analysis user to music ", comprises the steps：
All music all obtain 40 points of basis point 100；
Skin conductivity score 200：If xth song is in the label of " skin conductivity " this physiological data
" positive " is then obtained (25-100PSx) point, is that " the doubtful positive " then obtains (25-100PSx)/2 point,
0 point is then obtained for " feminine gender "；
Respiratory frequency score 300：If xth song is in the label of " respiratory frequency " this physiological data
" positive " is then obtained (20-100PRx) point, is that " the doubtful positive " then obtains (20-100PRx)/2 point,
0 point is then obtained for " feminine gender "；
Heart rate score 400：If xth song is " positive " in the label of " heart rate " this physiological data
Then obtain (15-100PHx) point, be that " the doubtful positive " then obtains (15-100PHx)/2 point, be " cloudy
Property " then obtain 0 point；
By " total score " 500=bases point 100+ skin conductivity score 200+ respiratory frequency score 300+ hearts
The formula of rate score 400 " calculates the first music total score 500 of xth.
Referring again to Fig. 5, on the other hand, wherein, " the physiology to different music and emotion response according to user
Degree, to user individualized music recommendation is carried out ", comprise the steps：
Step 1：User's xth (x >=2) it is secondary appreciate music when, with (x-1) secondary music total score press from
High to Low arrangement；
Step 2：The higher music of fraction, recommended frequency is higher, and fraction is lower, and recommended frequency is lower, often
The secondary new music for recommending a certain proportion of user never to listen, music arrangement mode is random；
Step 3：While user's xth (x >=2) secondary appreciation music, the physiological data of a new round is carried out
Collection and scoring；
Step 4：The scoring of all music of the secondary appreciation of user's xth (x >=2) is secondary with former (x-1)
Music total score is integrated, to live the music listened, sorting again after averaging, to what is do not listened in the past
Music, directly participates in sequence with the fraction that this is obtained；
Step 5：Xth (x >=2) wheel sequence is carried out to the music that user appreciates by score from big to small；
Step 6：Using xth wheel music ranking results as the personalized recommendation of ((x+1)th time) next time according to
According to.
Referring again to Fig. 6, a kind of personalized recommendation of music, including：
Physiological data collection module 600, for when user does not listen to music or listens to music, gathering user
Physiological data, while setting up " tranquillization physiological data storehouse " 610 and " action physiological data storehouse " 620；
Musical taste grading module 700, respectively to " tranquillization physiological data storehouse " 610 and " action physiology number
According to storehouse " 620 it is analyzed, obtain " the individualized music data base " 800 of user；
Individualized music recommending module 900, is carried out individual with reference to " individualized music data base " 800 couples of users
Property music recommend.
The above is only the preferred embodiment of the present invention, it is noted that common for the art
Technical staff, under the premise without departing from the principles of the invention, can also make some improvements and modifications, these
Improvements and modifications also should be regarded as protection scope of the present invention.
Claims (9)
1. a kind of personalized recommendation method of music, it is characterised in that comprise the steps：
Obtain user and listen to the every physiological data before music；
Obtain every physiological data during user's listening music；
By listening to music, obtain " feminine gender ", " positive " of items physiological data or " doubtful per a piece of music
Like positive " label；
Contrast user listens to the every physiological data before and after music, and analyzing user should to the physiology and emotion of music
Answer degree；
According to physiology and emotion response degree of the user to different music, to user individualized music recommendation is carried out.
2. the personalized recommendation method of music according to claim 1, it is characterised in that wherein, " obtain
Take family and listen to the physiological data before music ", comprise the steps：
With the bioimpedance data that bio-impedance sensor obtains user；
The physiological datas such as heart rate, respiratory frequency, the skin conductivity of user are calculated with the bioimpedance data for obtaining；
Before user listens to music, on an hourly basis the frequency of 6 times, gathers the physiology number under user's quiescent condition
According to：Heart rate samples, respiratory frequency sample and skin conductivity sample, achieve stand-by.
3. the personalized recommendation method of music according to claim 1, it is characterised in that wherein, " obtain
Take the physiological data during listening music of family ", comprise the steps：
With the bioimpedance data that bio-impedance sensor obtains user；
The physiological datas such as heart rate, respiratory frequency, the skin conductivity of user are calculated with the bioimpedance data for obtaining；
When user listens to music, by the frequency collection user of 6 times per minute xth (x is >=1 integer) is listened
Action physiological data during song：Heart rate samples, respiratory frequency sample and skin conductivity sample, prepare into
Row next step data analysiss.
4. the personalized recommendation method of music according to any one of Claims 2 or 3, it is special
Levy and be, the bio-impedance sensor can also be replaced using optoelectronic device.
5. the personalized recommendation method of music according to claim 1, it is characterised in that the life
Reason data include but is not limited to heart rate, breathing, skin conductivity, body temperature, blood pressure, blood oxygen, the muscle of user
The data such as tension force or biomolecule.
6. the personalized recommendation method of music according to claim 1, it is characterised in that the life
The score value of reason data includes but is not limited to 100 points of systems, ten point system or 1 point of system.
7. the personalized recommendation method of music according to claim 1, it is characterised in that wherein, " right
The every physiological data before and after music is listened to than user, physiology and emotion response journey of the user to music is analyzed
Degree ", comprises the steps：
All music all obtain 40 points of basis point；
Skin conductivity score：If xth song is " sun in the label of " skin conductivity " this physiological data
Property " (25-100PSx) point is then obtained, it is that " the doubtful positive " then obtains (25-100PSx)/2 point, it is " cloudy
Property " then obtain 0 point；
Respiratory frequency score：If xth song is " sun in the label of " respiratory frequency " this physiological data
Property " (20-100PRx) point is then obtained, it is that " the doubtful positive " then obtains (20-100PRx)/2 point, it is " cloudy
Property " then obtain 0 point；
Heart rate score：Obtain if xth song is " positive " in the label of " heart rate " this physiological data
(15-100PHx) point is obtained, is that " the doubtful positive " then obtains (15-100PHx)/2 point, be " feminine gender "
Then obtain 0 point；
By the public affairs on " total score=basis point+skin conductivity score+respiratory frequency score+heart rate score "
Formula calculates the first music total score of xth.
8. the personalized recommendation method of music according to claim 1, it is characterised in that wherein, " root
According to physiology and emotion response degree of the user to different music, to user individualized music recommendation is carried out ", bag
Include following steps：
User's xth (x >=2) is secondary appreciate music when, with (x-1) secondary music total score by arranging from high to low
Row；
The higher music of fraction, recommended frequency is higher, and fraction is lower, and recommended frequency is lower, recommends every time
The new music that a certain proportion of user never listened, music arrangement mode is random；
While user's xth (x >=2) secondary appreciation music, carry out the collection of the physiological data of a new round and comment
Point；
By the music total score that the scoring of all music of the secondary appreciation of user's xth (x >=2) is secondary with former (x-1)
Integrated, to live the music listened, sorting again after averaging, to the music do not listened in the past, directly
Connect the fraction obtained with this and participate in sequence；
Xth (x >=2) wheel sequence is carried out to the music that user appreciates by score from big to small；
Using xth wheel music ranking results as the personalized recommendation foundation of ((x+1)th time) next time.
9. the personalized recommendation of a kind of music, it is characterised in that include：
Physiological data collection module, for when user does not listen to music or listens to music, gathering the life of user
Reason data, while setting up " tranquillization physiological data storehouse " and " action physiological data storehouse "；
Musical taste grading module, enters respectively to " tranquillization physiological data storehouse " and " action physiological data storehouse "
Row analysis, obtains " the individualized music data base " of user；
Individualized music recommending module, individualized music is carried out with reference to " individualized music data base " to user
Recommend.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510708821.5A CN106611028A (en) | 2015-10-27 | 2015-10-27 | Personalized music recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510708821.5A CN106611028A (en) | 2015-10-27 | 2015-10-27 | Personalized music recommendation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106611028A true CN106611028A (en) | 2017-05-03 |
Family
ID=58614624
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510708821.5A Pending CN106611028A (en) | 2015-10-27 | 2015-10-27 | Personalized music recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106611028A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108882045A (en) * | 2017-05-11 | 2018-11-23 | 昆山研达电脑科技有限公司 | A kind of film scoring apparatus and method based on wearable device |
| CN109271549A (en) * | 2018-09-30 | 2019-01-25 | 百度在线网络技术（北京）有限公司 | Song recommendations method, apparatus, terminal and computer readable storage medium |
| CN109636442A (en) * | 2018-11-12 | 2019-04-16 | 咪咕文化科技有限公司 | Method, device and storage medium for determining favorite contents |
| CN110096613A (en) * | 2019-04-12 | 2019-08-06 | 北京奇艺世纪科技有限公司 | A kind of video recommendation method, device, electronic equipment and storage medium |
| CN110134822A (en) * | 2019-05-10 | 2019-08-16 | 一汽轿车股份有限公司 | Vehicle-mounted music recommended method and recommender system |
| CN114470475A (en) * | 2022-01-25 | 2022-05-13 | 福州亚历山大健康管理有限公司 | Scene type cognitive guide sound wave treatment system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101934111A (en) * | 2010-09-10 | 2011-01-05 | 李隆 | Music chromatic light physical factor physical and mental health system based on computer |
| CN102523493A (en) * | 2011-12-09 | 2012-06-27 | 深圳Tcl新技术有限公司 | Method and system for grading television program according to mood |
| CN104123355A (en) * | 2014-07-17 | 2014-10-29 | 深圳市明康迈软件有限公司 | Music recommendation method and system |
| CN104809221A (en) * | 2015-04-30 | 2015-07-29 | 努比亚技术有限公司 | Recommending method for music information and device |
| US20150297109A1 (en) * | 2014-04-22 | 2015-10-22 | Interaxon Inc. | System and method for associating music with brain-state data |
-
2015
- 2015-10-27 CN CN201510708821.5A patent/CN106611028A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101934111A (en) * | 2010-09-10 | 2011-01-05 | 李隆 | Music chromatic light physical factor physical and mental health system based on computer |
| CN102523493A (en) * | 2011-12-09 | 2012-06-27 | 深圳Tcl新技术有限公司 | Method and system for grading television program according to mood |
| US20150297109A1 (en) * | 2014-04-22 | 2015-10-22 | Interaxon Inc. | System and method for associating music with brain-state data |
| CN104123355A (en) * | 2014-07-17 | 2014-10-29 | 深圳市明康迈软件有限公司 | Music recommendation method and system |
| CN104809221A (en) * | 2015-04-30 | 2015-07-29 | 努比亚技术有限公司 | Recommending method for music information and device |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108882045A (en) * | 2017-05-11 | 2018-11-23 | 昆山研达电脑科技有限公司 | A kind of film scoring apparatus and method based on wearable device |
| CN109271549A (en) * | 2018-09-30 | 2019-01-25 | 百度在线网络技术（北京）有限公司 | Song recommendations method, apparatus, terminal and computer readable storage medium |
| CN109636442A (en) * | 2018-11-12 | 2019-04-16 | 咪咕文化科技有限公司 | Method, device and storage medium for determining favorite contents |
| CN110096613A (en) * | 2019-04-12 | 2019-08-06 | 北京奇艺世纪科技有限公司 | A kind of video recommendation method, device, electronic equipment and storage medium |
| CN110096613B (en) * | 2019-04-12 | 2021-07-20 | 北京奇艺世纪科技有限公司 | Video recommendation method and device, electronic equipment and storage medium |
| CN110134822A (en) * | 2019-05-10 | 2019-08-16 | 一汽轿车股份有限公司 | Vehicle-mounted music recommended method and recommender system |
| CN114470475A (en) * | 2022-01-25 | 2022-05-13 | 福州亚历山大健康管理有限公司 | Scene type cognitive guide sound wave treatment system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106611028A (en) | Personalized music recommendation method and system | |
| Salai et al. | Stress detection using low cost heart rate sensors | |
| Jaimovich et al. | Emotion in motion: A study of music and affective response | |
| Ellamil et al. | One in the dance: musical correlates of group synchrony in a real-world club environment | |
| CN107066801A (en) | Method and system for analyzing sound | |
| Monahan et al. | Auditory sensitivity to formant ratios: Toward an account of vowel normalisation | |
| Graham et al. | Patterns and predictors of tobacco consumption among women | |
| Yu et al. | A hybrid user experience evaluation method for mobile games | |
| Seo et al. | Emotional competence and online game use in adolescents | |
| Tervaniemi et al. | Importance of the left auditory areas in chord discrimination in music experts as demonstrated by MEG | |
| Rodero et al. | Synthetic versus human voices in audiobooks: The human emotional intimacy effect | |
| Bari et al. | rconverse: Moment by moment conversation detection using a mobile respiration sensor | |
| Stepp et al. | Modulation of neck intermuscular beta coherence during voice and speech production | |
| CN107767965A (en) | Health monitoring system and method for multi-factor correlation comparison | |
| Morgan | Categorical and dimensional ratings of emotional speech: Behavioral findings from the Morgan emotional speech set | |
| Chao et al. | Boy crisis? Sex differences in self-injurious behaviors and the effects of gender role conflicts among college students in China | |
| Gashi et al. | A multidevice and multimodal dataset for human energy expenditure estimation using wearable devices | |
| AU2021101097A4 (en) | A system and method for automatic playlist generation by analysing human emotions through physiological signals | |
| Li et al. | Exploration of combined physical activity and music for patients with Alzheimer’s disease: A systematic review | |
| KR20150000591A (en) | System and method for lifestyle analysis | |
| Talebi et al. | Effects of vowel auditory training on concurrent speech segregation in hearing impaired children | |
| An et al. | Mental health detection from speech signal: A convolution neural networks approach | |
| Rodrigues et al. | Questionnaires measuring 24-hour movement behaviors in childhood and adolescence: Content description and measurement properties—a systematic review | |
| Düvel et al. | Experience of groove questionnaire: German translation and validation | |
| Lee et al. | Music for sleep and wake-up: an empirical study |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170503 |