Music personalized recommendation method based on physiological features

Abstract

The invention discloses a music personalized recommendation method based on physiological features, and the method comprises the steps: obtaining the physiological feature data of a wearer based on a portable device, and transmitting the physiological feature data to interaction equipment; based on the physiological feature data, acquiring the emotion of the wearer; the interaction device obtains the music comment and obtains an emotion mobilization parameter corresponding to the music; providing expected recommended music for the wearer on the basis of the emotion transfer parameter and the emotion parameter of the wearer; acquiring physiological feature data of the wearer when the wearer listens to the expected recommended music, and acquiring an emotion matching degree of the expected recommended music; inputting the music of which the emotion matching degree is not lower than a preset emotion matching degree into a recommended music database; and on the basis of the emotion parameters of the wearer, supplementing available expected recommended music in a non-recommended music database into the recommended music database so as to carry out personalized recommendation of the music. The problem that physiological features are not considered in music recommendation is solved, and therefore the emotion of the person can be adjusted based on music personalized recommendation.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and in particular relates to a personalized music recommendation method based on physiological characteristics. Background Art

[0002] Various music platforms currently have developed different personalized music recommendation methods. These methods primarily determine the user's preferred music genre based on their historical listening habits, and then make personalized recommendations based on that genre. However, human emotions are easily affected by music. Furthermore, when a user experiences a long-term negative mood, the music they listen to is likely to also contain negative emotions. If similar music styles are constantly recommended, this can exacerbate the user's negative emotions, impacting their physical health over time. Furthermore, long-term listening to a single style of music can easily lead to aesthetic fatigue in music appreciation, hindering the user's appreciation of other types of music and creating hidden stress for the user when listening to music. Music itself can become a component of negative emotions. In other words, current personalized music recommendations can easily exacerbate users' negative emotions, to the point where music itself can become the basis for negative emotions.

[0003] Therefore, how to adjust or maintain the user's emotions based on physiological characteristics is a technical problem that those skilled in the art urgently need to solve. Summary of the Invention

[0004] In order to utilize personalized music recommendation solutions to help users adjust their negative emotions or maintain their positive emotions, this application discloses a personalized music recommendation method based on physiological characteristics, specifically: A personalized music recommendation method based on physiological characteristics, the recommendation method comprising: Based on the portable device, the wearer's physiological characteristic data is obtained and sent to the interactive device; The interactive device acquires the wearer's emotions based on the physiological characteristic data to obtain the wearer's emotional parameters; The interactive device obtains music reviews and obtains emotion mobilization parameters corresponding to the music; providing the wearer with expected recommended music based on the emotion mobilization parameter and the wearer's emotion parameter; Acquiring physiological characteristic data of the wearer when listening to the expected recommended music, and obtaining the emotional matching degree of the expected recommended music; Inputting the music with an emotion matching degree not less than a preset emotion matching degree into a recommended music database, and associating the recommended music database with the wearer's emotion parameter; Based on the wearer's emotional parameters, the available expected recommended music in the non-recommended music database is added to the recommended music database to perform personalized music recommendations.

[0005] Optionally, the method of acquiring the wearer's physiological characteristic data based on the portable device and sending the data to the interactive device includes: The portable device continuously monitors physiological characteristics of the wearer and generates physiological characteristic data; A communication relationship is established between the portable device and the interactive equipment, and the physiological characteristic data is sent to the interactive equipment.

[0006] Optionally, the interactive device acquires the wearer's emotions based on the physiological characteristic data to obtain the wearer's emotional parameters, including: Acquiring the physiological characteristic data and removing interference items to obtain usable physiological characteristic data; Acquire mutual influences between the available physiological characteristic data, obtain correlation relationships between the available physiological characteristic data, and acquire an available physiological characteristic data group based on the correlation relationships; Selecting one available physiological characteristic data from each of the available physiological characteristic data groups, and obtaining all available physiological characteristic data combinations based on the available physiological characteristic data groups to construct an emotion parameter analysis group; Based on the emotional parameter analysis groups, the wearer's emotional parameters corresponding to each of the emotional parameter analysis groups are obtained.

[0007] Optionally, before determining the wearer's emotional parameter, the method further includes: Setting an evaluation time step for the physiological characteristic data; Based on the evaluation time step, performing a fluctuation measurement on the physiological characteristic data to obtain a fluctuation amplitude; Acquiring physiological characteristic data whose fluctuation amplitude is not less than a preset fluctuation amplitude, and acquiring a time point at which the fluctuation of the physiological characteristic data occurs; Obtaining physiological characteristic data within each evaluation time step after the time node at which the fluctuation occurs, obtaining the fluctuation amplitude of the physiological characteristic data within adjacent evaluation steps, and obtaining adjacent fluctuation amplitudes; Comparing the adjacent fluctuation amplitudes with preset adjacent fluctuation amplitudes, and obtaining a duration in which the adjacent fluctuation amplitudes are not higher than the preset adjacent fluctuation amplitudes; When the duration is not less than the preset duration, it is determined that the wearer's emotional parameter needs to be obtained.

[0008] Optionally, the interactive device obtains music reviews and obtains emotion mobilization parameters corresponding to the music, including: Obtain music reviews from the music comment section and obtain all emotional words in the music reviews; Based on the association relationship between all the emotion words, obtaining the logical relationship between all the emotion words; Obtain all the logical relationships in the music review, and obtain the number of times all the logical relationships are generated; Based on the number of times generated, obtaining an emotion mobilization parameter of the music corresponding to the music review; The music with the same or similar emotion-mobilizing parameters is added to the music library corresponding to the emotion-mobilizing parameters to obtain an emotion-mobilizing music library.

[0009] Optionally, providing the wearer with expected recommended music based on the emotion mobilization parameter and the wearer's emotion parameter includes: Obtaining a matching degree between the emotion mobilization parameter and the wearer's emotion parameter; Sorting the matching degrees to obtain a matching degree priority sequence; Based on the matching priority sequence, obtaining the emotion-arousing music library at the top of the matching priority sequence; Based on the matching ratio, randomly selecting music from all the emotion-arousing music libraries according to the ratio, and setting the music into the expected recommendation database; The music recommended to the wearer is randomly determined from the expected recommendation database, and the recommended music is the expected recommended music.

[0010] Optionally, obtaining physiological characteristic data of the wearer when listening to the expected recommended music and obtaining the emotional matching degree of the expected recommended music includes: Continuously acquiring physiological characteristic data of the wearer while the wearer listens to the expected recommended music to obtain changed physiological characteristic data; Obtaining a change value of the changed physiological characteristic data to obtain a time starting point of the emotion change; At the starting point of the emotion change, obtaining the changed emotion parameter; The changed emotional parameter and the expected recommendation database matching degree corresponding to the expected recommended music are obtained to obtain the emotional matching degree of the expected recommended music.

[0011] Optionally, the step of inputting the music having an emotion matching degree not less than a preset emotion matching degree into a recommended music database, and associating the recommended music database with the wearer's emotion parameter, includes: Setting a preset emotion matching degree to determine the matching degree between the expected recommended music and the wearer's emotion parameters; Acquiring physiological characteristic data of the wearer during a listening period of a single piece of the expected recommended music, and acquiring emotional parameters during a listening period of a single piece of music; Obtaining a matching degree between the emotion parameter during the listening period of the single piece of music and the emotion mobilization parameter to obtain an emotion matching degree; Acquire music whose emotion matching degree is not less than a preset emotion matching degree, and input the music into a recommended music database; The recommended music database is associated with the wearer's emotional parameters during listening to a single piece of music, so as to determine the recommended music database based on the wearer's emotional parameters.

[0012] Optionally, the process of acquiring the emotional parameters during listening to a single piece of music further includes: During a single period of the expected recommended music recommendation, setting an evaluation time step of the music listening process; Obtaining the emotional matching degree of all single music listening periods based on the evaluation time step of the music listening process; Obtaining a deviation between the emotion matching degree and the preset emotion matching degree, and constructing a deviation curve; Based on the deviation curve, obtaining the wearer's mood change time point, the mood change time point being the starting point of the wearer's mood change in the next stage; Obtain the wearer's emotions in the next stage and build the expected recommended music library for the next stage; When the starting point of the wearer's mood change in the next stage is reached, music in the expected recommendation database for the next stage is recommended to the wearer.

[0013] Optionally, based on the wearer's emotional parameter, the method of adding the available expected recommended music in the non-recommended music database to the recommended music database to perform personalized music recommendation includes: Based on the wearer's emotional parameter and the preset emotional matching degree, obtaining music that matches the wearer's emotional parameter to obtain available expected recommended music; Comparing the available expected recommended music with the music already in the recommended music database, obtaining the available expected recommended music in the non-recommended music library, and adding the music to the recommended music database; The music in the recommended music database is randomly sorted and recommended to the wearer.

[0014] The beneficial effects of this application include: 1. Personalized music recommendations based on physiological characteristics are realized. In the technical solution of this application, the recommended music for users is no longer processed based on characteristics such as music style and type. Instead, the user's physiological characteristics are obtained, and then the emotional adjustment requirements are determined based on the collected physiological characteristic data. At the same time, the emotional description of the music comment area is obtained to determine the emotional expression of the music itself. Based on this method, it is more in line with the actual emotional expression of the music listeners. And based on the user's emotional adjustment requirements and the actual emotional expression of the music itself, it is truly possible to adjust or maintain the user's emotions based on music recommendations.

[0015] 2. Improved rationality of personalized music recommendations. The technical solution of this application determines the user's emotional fluctuations through real-time monitoring of the user's physiological characteristics and analysis of the fluctuation rate of physiological parameters. It further analyzes the user's specific emotions and selects music recommendations based on the user's emotions. The recommended music matches the user's emotional adjustment or maintenance needs, thereby improving the rationality of personalized music recommendations.

[0016] 3. Expanding the scope of personalized music recommendations. The technical solution of this application, after creating a recommended music database, also obtains other music with the same emotion based on the user's emotional parameters corresponding to the database, adds it to the music database, and recommends it to the user. Because each recommended music is obtained based on the user's physiological characteristics, it can cover a variety of music styles and types, effectively expanding the music included in the music database. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the embodiments of the present application or the prior art. Obviously, the following description is only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. The drawings are used to provide a further understanding of the present disclosure and constitute part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure, but do not constitute a limitation of the present disclosure. In the drawings: Figure 1 A flowchart of a personalized music recommendation method based on physiological characteristics provided in an embodiment of the present application; Figure 2 This is a deviation curve of a personalized music recommendation method based on physiological characteristics provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. In addition, in the embodiments of the present application, "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0019] Currently, people of all kinds are paying more attention to their personal health. To this end, they wear smart watches, bracelets, and other portable health monitoring devices in their daily work and life. These devices can obtain the wearer's physiological characteristic data, such as blood pressure, heart rate, blood oxygen, body temperature, sleep time, sitting time, and even stress. However, this method can only determine the sub-health status faced by an individual in the past period of time and cannot provide effective advice based on this physiological information. At the same time, a large number of people currently use music to adjust their living and working environment at work and in life. Considering that music naturally has the effect of adjusting, maintaining, and even enhancing human emotions, and this effect will also have a reverse effect on physiological characteristics, it can be said that adjusting sub-health status based on music is feasible. However, in current personalized music recommendations, reasonable personalized music recommendations based on the physiological characteristic data obtained by portable health monitoring devices are not considered to adjust or maintain the user's mood. Therefore, the purpose of this application is to monitor the user's physiological characteristic data and realize reasonable personalized music recommendations based on the obtained data, thereby adjusting the user's physiological characteristic indicators.

[0020] This application discloses a personalized music recommendation method based on physiological characteristics, such as Figure 1 FIG. 1 is a flowchart of a method for personalized music recommendation based on physiological characteristics provided in an embodiment of the present application, including: S110 , obtaining the wearer's physiological characteristic data based on the portable device and sending it to the interactive device.

[0021] S120: The interactive device obtains the wearer's emotions based on the physiological characteristic data to obtain the wearer's emotional parameters.

[0022] S130: The interactive device obtains music reviews and obtains emotion mobilization parameters corresponding to the music.

[0023] S140: Providing the wearer with expected recommended music based on the emotion mobilization parameter and the wearer's emotion parameter.

[0024] S150: Acquire physiological characteristic data of the wearer when listening to the expected recommended music, and acquire the emotional matching degree of the expected recommended music.

[0025] S160: Input the music whose emotion matching degree is not less than a preset emotion matching degree into a recommended music database, and associate the recommended music database with the wearer's emotion parameter.

[0026] S170: Based on the wearer's emotional parameters, the available expected recommended music in the non-recommended music database is added to the recommended music database to perform personalized music recommendation.

[0027] The purpose of all the above steps is to ensure that the recommended music matches the mood of the wearer of the portable device, and based on the appropriate recommended music, the wearer's emotions are appropriately affected, such as reducing negative emotions and maintaining or enhancing positive emotions. At the same time, the wearer's emotional changes are also taken into account, thereby forming a continuous and reasonable emotional adjustment mechanism, while expanding the range of music in the recommendation database to provide the wearer with the best personalized music recommendation plan.

[0028] Below, all the above steps will be described in detail, specifically: As described in step S110, the purpose of this step is to obtain the current physiological characteristics of the music listener and process the data by the corresponding music providing device, that is, the interactive device, so as to lay the foundation for the subsequent personalized music recommendation work. Specifically: S111. The portable device continuously monitors the wearer's physiological characteristics and generates physiological characteristic data.

[0029] The purpose of this step is to analyze the emotions of the wearer of the portable device. In order to improve the rationality and accuracy of personalized music recommendations, it is obviously necessary to quantify the emotions. Therefore, this step can lay the foundation for the quantitative analysis process by continuously monitoring the wearer's physiological characteristics and generating numerical values.

[0030] Among them, the portable device is a device that can measure the wearer's physiological characteristics. At the same time, the device must also have communication functions, including smart watches, bracelets, some smart medical devices, etc., which are not limited in this application.

[0031] The portable device continuously obtains the physiological characteristics of the wearer and measures the corresponding physiological characteristic parameters to obtain physiological characteristic data.

[0032] Among them, physiological characteristic data include heart rate, blood oxygen, blood pressure, body temperature, sleep time, sitting time, and physiological or psychological stress, etc. Obviously, other physiological characteristic data can also be included. As long as the data can be measured by the portable device, it can be used as physiological characteristic data, and this application does not limit it.

[0033] S112: Establish a communication relationship between the portable device and the interactive device, and send the physiological characteristic data to the interactive device.

[0034] The purpose of this step is that it is obviously unrealistic to directly process the data obtained by the portable device. At the same time, the portable device often does not have a music playback function, so the obtained data needs to be sent to the interactive device, and the interactive device completes the data processing.

[0035] Among them, the interactive device needs to have a communication relationship and be able to communicate with the portable device.

[0036] Among them, interactive devices include mobile phones, personal PC devices and various music apps. Of course, it also supports the setting of signal relay devices between portable devices and interactive devices. For example: the wearer wears a portable device and uses a personal PC device to play music at the same time. The portable device cannot be directly connected to the personal PC device. At this time, the portable device can be connected to the mobile phone, and the mobile phone can be connected to the personal PC device. At this time, the processing of physiological characteristic data can be processed by the personal PC device and / or mobile phone, and this application does not limit it.

[0037] The beneficial effect of S110 is that it can realize accurate and real-time monitoring of the wearer's physiological characteristic data, and at the same time the interactive device processes the data, realizing professional processing of the physiological characteristic data.

[0038] As described in step S120, the purpose of this step is to process the wearer's current emotions, so as to determine the wearer's specific emotions based on the obtained physiological characteristic data. Specifically: S121. Acquire the physiological characteristic data, and remove interference items to obtain usable physiological characteristic data.

[0039] The purpose of this step is to pre-process the physiological characteristic data currently collected by the wearer to ensure that the quality of the acquired physiological characteristic data can meet the subsequent use requirements. Specifically: Among them, for the obtained physiological characteristic data, data with excessive fluctuations in the physiological characteristic data are obtained, and such data are considered as interference items and are eliminated.

[0040] When eliminating interference items from the obtained physiological characteristic data, it is necessary to shorten the interference item recognition time of the physiological characteristic data to the greatest extent possible to avoid considering situations with large emotional fluctuations as interference items.

[0041] This also includes identifying and removing other types of interference items that can be identified.

[0042] S122: Obtain mutual influences between the available physiological characteristic data, obtain correlation relationships between the available physiological characteristic data, and obtain an available physiological characteristic data group based on the correlation relationships.

[0043] The purpose of this step is to identify some physiological characteristics that are correlated with each other, such as heart rate is usually correlated with blood pressure. At the same time, it can also be determined based on a physiological characteristic and other physiological characteristics that cannot be detected, such as blood oxygen content may be correlated with breathing depth and frequency. In order to better determine the wearer's emotions, it is necessary to obtain physiological characteristic data that do not have any correlation. Specifically: Among them, for all physiological characteristic data that can be collected by the portable device, the mutual correlation between such physiological characteristic data is determined, and their specific correlation is determined.

[0044] In some embodiments, each physiological characteristic data collected needs to be judged. The specific judgment method is: obtain the research results of the correlation between the currently available physiological characteristic data, and at the same time compare the changes and correlation research results of the same physiological characteristic data. If the two are found to be different, it is considered that there is no correlation between the two physiological characteristic data.

[0045] Among them, all physiological characteristic data without any correlation relationship are classified and arranged. Specifically, they are arranged in different databases. For other physiological characteristic data with correlation relationship, they are distributed and arranged in the same database, thus generating a usable physiological characteristic data group. For example, in the acquisition of physiological characteristic data, the correlation obtained is shown in Table 1: ; Among them, after analysis, it was found that there is a correlation between the obtained blood pressure and heart rate. This correlation usually occurs during anaerobic exercise. However, for blood oxygen saturation and body temperature, although the performance is similar to the possible data during anaerobic exercise, the data is not yet related to the exercise state. Therefore, these two data are used as separate data groups. Regarding sleep time, it is obviously an independent physiological characteristic data, so an independent data group is set up.

[0046] Among them, the correlation between physiological characteristic data can be determined based on various situations such as scenes (such as sports scenes, learning scenes, sleep scenes, etc.), movement states (such as uniform speed movement, variable speed movement, etc.), etc., and this application does not limit it.

[0047] S123: Select one available physiological characteristic data from each of the available physiological characteristic data groups, and obtain all available physiological characteristic data combinations based on the available physiological characteristic data groups to construct an emotion parameter analysis group.

[0048] The purpose of this step is to obtain the parameters of each data group in the available physiological characteristic data group, and then process the data in each data group to obtain the emotion parameter analysis group. Specifically: A type of physiological characteristic data is obtained from each different physiological characteristic data group.

[0049] The obtained physiological characteristic data is constructed into a corresponding combination, and the physiological characteristic type is obtained. For example, for the data obtained in Table 1, the data combination established is {blood pressure, blood oxygen saturation; body temperature; sleep time} or {heart rate; blood oxygen saturation; body temperature; sleep time}, and the measured physiological characteristic data is added to the array.

[0050] The obtained array is the emotion parameter analysis group.

[0051] S124. Based on the emotion parameter analysis groups, obtain the wearer's emotion parameters corresponding to each emotion parameter analysis group.

[0052] The purpose of this step is to determine the wearer's emotional parameters based on the corresponding physiological characteristic data in the data set after obtaining the emotional parameter data set. Specifically: The change rate of each physiological characteristic data is obtained.

[0053] The change rate of all physiological characteristic data is added to the physiological characteristic change equation to determine the current physiological characteristic change situation. The emotion fluctuation judgment equation is: ; in, Indicates the rate of change of physiological characteristics; Indicates the i The rate of change of physiological characteristic data in each emotional parameter data group; i The index of the sentiment parameter analysis group; j Indicates the total number of indexes in the sentiment parameter analysis group.

[0054] After determining the physiological characteristic change rate, the ratio of the parameter to each physiological characteristic data is determined, and then the data amplitude of the current physiological characteristic data change is analyzed. The equation is: ; in, Indicates the i The change weight of the physiological characteristic data of each emotion parameter data group.

[0055] Considering that different emotions produce different changes in physiological characteristic data, for example, when excited, blood oxygen levels rise most significantly, while when depressed, all physiological characteristic data decrease. When frightened, a "flight response" occurs, with heart rate, blood oxygen levels, and blood pressure all rising simultaneously and significantly. Therefore, the wearer's emotional parameters can be determined based on the weight of the resulting changes.

[0056] The judgment of the emotional parameters may also be based on other algorithms and currently studied information, and this application does not limit this.

[0057] Among them, considering that in some scenarios, the wearer's emotions themselves remain unchanged, but their physiological characteristic data will change. For example, during aerobic exercise, the wearer's emotions remain stable, but at this time it is necessary to apply incentives to the wearer or reduce the physiological characteristic data to avoid the mistaken transition from aerobic to anaerobic. To cope with this situation, this application also considers introducing a scene judgment mechanism, specifically: Among them, a change curve of the wearer's physiological characteristic data is established. When it is in a stable growth state, it is considered that it may be in a gradually active state.

[0058] In some embodiments, positioning information, such as Beidou navigation information, GPS location information, etc., can also be introduced to determine whether the person is in motion.

[0059] When the wearer is in a sports scene, the wearer's specific emotions are not considered, but the physiological characteristics data is used as the basis for reduction. In other words, for some scenes such as sports, these scenes are given the highest priority for music recommendation, regardless of the wearer's specific emotions.

[0060] The beneficial effect of this step is that when determining the emotional parameters by using the data in the emotional parameter analysis group, the emotional parameters of the wearer are accurately determined by analyzing information such as the wearer's scene, the rate of change of physiological characteristic data, and the change weights based on various physiological characteristic data. At the same time, it can cope with various scenarios and jointly determine subsequent music recommendation plans based on the scenarios and emotional parameters.

[0061] Based on step S120, the wearer's mood can be determined. However, considering that the purpose of this application is to adjust the wearer's mood throughout the entire process, or to maintain the wearer's mood throughout the entire process in order to adjust the physiological characteristic data, it is obviously necessary to obtain the wearer's physiological characteristic data throughout the entire process and to reasonably process the data that can be obtained. Therefore, before determining the wearer's emotional parameters, the following is also included: S12 (1) setting an evaluation time step for the physiological characteristic data.

[0062] The purpose of this step is to determine the time range for processing the obtained physiological characteristic data, and then collect data based on this time range. Specifically: Among them, the evaluation time step needs to be reasonably lowered to ensure the accuracy of subsequent data processing, for example, 10s is used as an evaluation time step.

[0063] S12 (2) Based on the evaluation time step, the fluctuation of the physiological characteristic data is measured to obtain the fluctuation amplitude.

[0064] The purpose of this step is that when the wearer experiences emotional fluctuations, it usually indicates that the wearer's situation or personal state has changed, and the wearer needs to adjust or maintain the emotional state after the fluctuation. The basis of this technology is to obtain the emotional fluctuations. Considering that the physiological characteristic data usually changes accordingly after emotional fluctuations, it is feasible to obtain the volatility of physiological characteristic data to judge emotional fluctuations. Specifically: The fluctuation situation is obtained by taking the mean of each physiological characteristic data within each evaluation time step.

[0065] In some embodiments, it is also possible to only measure the corresponding parameters of the physiological characteristic data at the starting point of each evaluation time step, and calculate the change rate between two adjacent data. This parameter is the fluctuation rate.

[0066] S12 (3), obtaining physiological characteristic data with a fluctuation amplitude not less than a preset fluctuation amplitude, and obtaining a time point at which the physiological characteristic data fluctuates.

[0067] The purpose of this step is to determine whether the wearer's mood changes only slightly and briefly during their current activity. Considering that most current music lasts for 3 to 5 minutes, it is obviously not necessary to adjust to such mood fluctuations by immediately switching music. Therefore, in such cases, it is necessary to determine whether the current fluctuation amplitude requires intervention and adjustment. Specifically: The preset fluctuation amplitude may be set based on the adjustment sensitivity requirement of the emotion.

[0068] In some embodiments, the setting can be made based on changes in physiological characteristic data during the wearer's historical emotional expressions.

[0069] The obtained fluctuation amplitude is compared with the preset fluctuation amplitude. If the former is not lower than the latter, it is considered that the current emotional fluctuation is in a state that needs to be adjusted.

[0070] Each time an emotional fluctuation that needs to be adjusted occurs, it is necessary to obtain the time node of the current large emotional fluctuation.

[0071] S12 (4) obtains the physiological characteristic data within each evaluation time step after the time node where the fluctuation occurs, obtains the fluctuation amplitude of the physiological characteristic data within adjacent evaluation steps, and obtains the adjacent fluctuation amplitude.

[0072] The purpose of this step is to understand that in some cases, the wearer's mood may fluctuate greatly in a short period of time, but overall the mood is stable. In this case, it is difficult to adjust the mood based on music. Therefore, it is necessary to analyze whether the wearer's mood fluctuations are long-term, because only long-term mood fluctuations can be adjusted by music. Among them, when the emotional fluctuation is determined, the time point when it occurs is the starting time point for further evaluation.

[0073] Among them, the physiological characteristic data within adjacent evaluation steps are determined, and then the fluctuation amplitude of the physiological characteristic data within two adjacent evaluation steps is analyzed, and the obtained result is the adjacent fluctuation amplitude.

[0074] S12 (5), comparing the adjacent fluctuation amplitudes with the preset adjacent fluctuation amplitudes, and obtaining the duration during which the adjacent fluctuation amplitudes are not higher than the preset adjacent fluctuation amplitudes.

[0075] The purpose of this step is that if the wearer's adjacent fluctuations are small within adjacent time steps, it can be considered that the wearer's emotions remain stable within these two time steps. Based on this common sense, it can be used to determine whether the current wearer needs to adjust his or her emotions. Specifically: The preset adjacent fluctuation amplitudes may be set based on research results of emotion changes or emotion adjustment accuracy.

[0076] In some embodiments, the preset adjacent fluctuation amplitudes may be determined based on the wearer's historical physiological data.

[0077] The adjacent fluctuation amplitudes are compared with the preset adjacent fluctuation amplitudes. When it is found that the adjacent fluctuation amplitudes are not higher than the preset adjacent fluctuation amplitudes, it is determined that the wearer's emotions in the sequences of the two evaluation time steps remain stable.

[0078] Among them, when obtaining the adjacent fluctuation amplitude and comparing it with the preset adjacent fluctuation amplitude, the evaluation time step of each adjacent fluctuation is obtained, and then based on the number of evaluation time steps, the duration of the adjacent fluctuation amplitude not higher than the preset adjacent fluctuation amplitude state is determined.

[0079] S12 (6): When the duration is not less than the preset duration, it is determined that the wearer's emotional parameters need to be obtained.

[0080] The purpose of this step is that when the wearer's emotions need to be adjusted or maintained, the adjacent fluctuation amplitude of the person is small, so based on this method, it can be determined whether the wearer's emotions need to be adjusted.

[0081] The preset duration can be set based on the time it takes for various emotions to be generated and calmed down.

[0082] The parameter may also be set based on the wearer's historical physiological characteristic data.

[0083] When it is found that the duration is not less than the preset duration, it is considered necessary to determine the wearer's emotional parameters based on the method of step S120.

[0084] The beneficial effect of this step is that it is equivalent to decomposing the wearer's emotional changes and obtaining adjacent data fluctuations to determine whether the current emotion needs to be adjusted. In addition, for the generation and maintenance process of emotions, the enhancement or maintenance of emotions is usually in a state where the changes are relatively fixed in a short period of time and the degree of change is very small. Therefore, by analyzing the fluctuation amplitude of physiological characteristic data within adjacent evaluation time steps, it can adapt to all types of emotional changes. At the same time, it can also remove situations where emotions fluctuate for a very short time (that is, situations where it is difficult to make emotional fluctuations based on the music itself). This not only improves the accuracy of emotional adjustment, but also ensures that the entire technical solution can be implemented, thus being practical.

[0085] As described in step S130, the purpose of this step is that for various types of music, they themselves carry information labels such as style and author. However, this information label is not equivalent to the emotions contained in the music itself. However, for listeners, they often have emotional thoughts while listening to music. Therefore, based on the music emotion vocabulary in the comment area, it is used to ensure high-precision analysis of the emotions carried by the music. Specifically: S131. Obtain music reviews in a music comment area, and obtain all emotional words in the music reviews.

[0086] The purpose of this step is to analyze the emotional words in the music comment area and lay the foundation for the music's emotional analysis process.

[0087] Among them, the comment area of ​​the music is determined, and various emotional words are retrieved from it.

[0088] Among them, conjunctions and transition words can also be included to accurately determine the most critical emotional words in the music.

[0089] S132: Based on the association relationship between all the emotion words, obtain the logical relationship of all the emotion words.

[0090] The purpose of this step is that for some comments, there may be various relationships between the emotional words, such as transition, parallel, and advancement, but it is difficult to judge based on the emotional words themselves. Therefore, it is necessary to obtain logical relationships based on the random combinations of various emotional words obtained, and then further determine the specific emotions based on the logical relationships.

[0091] Among them, all emotion words are randomly arranged and combined, including two-word combinations, three-word combinations, and up to n-word combinations.

[0092] Among them, for the combination of emotional words, the meaning of the words in each combination is analyzed. If the meanings are the same or similar, the combination is identified as one word.

[0093] Among them, all the individual words are further combined to establish a relationship, and it is judged whether there are any emotional transitions, parallelism, advancement and other connection relationships between the various words in the established combination relationship.

[0094] After obtaining the connection relationship, the specific logical relationship between the various words is determined. The so-called logical relationship refers to the way in which the emotional words are related to each other, including transitions, parallelism, etc.

[0095] S133. Obtain all the logical relationships in the music review, and obtain the number of times all the logical relationships are generated.

[0096] The purpose of this step is to obtain the logical relationship. Not all comments have the corresponding logical relationship. At the same time, the number of times such logical relationships are generated can be used to conduct a concise analysis of the music reviews.

[0097] After the logical relationship is obtained, the meaning of the specific emotion words in the logical relationship is determined, and based on the logical relationship, all similar emotion words are combined into vocabulary according to the logical relationship.

[0098] Among them, based on the vocabulary combination that has been established, all the comments on the music are indexed based on the logical relationship of the vocabulary combination. If it is obtained through the index, it is considered that a logical relationship appears in the comment.

[0099] Among them, for some long comments, it is necessary to index the entire comment and analyze the number of occurrences of logical relationships.

[0100] The number of times all logics appear in the comments is recorded, thereby obtaining the number of times they are generated.

[0101] Among them, the logical relationship generated by each emotional word is searched separately and the generation count is obtained. Then, the generation count of the logical relationships with the same or similar meanings of the emotional words is added together to obtain the specific generation count of a certain logical relationship.

[0102] S134. Based on the number of times the music review is generated, an emotion mobilization parameter of the music corresponding to the music review is obtained.

[0103] The purpose of this step is that when a certain logical relationship is generated the most times, it is obvious that the emotional adjustment ability corresponding to this logical relationship is the highest, because more listeners have this idea. Obviously, it is feasible to determine the emotional mobilization ability based on this method.

[0104] The obtained number of times of generation is sorted, and the obtained sorting is directly used as an emotion mobilization parameter.

[0105] In some embodiments, the recognition of music reviews is considered. Specifically, after obtaining a comment with a corresponding logical relationship, the number of approvals for the comment is determined, and the ratio of the number of approvals to the total number of approvals is calculated. This ratio can be directly used as the recognition, and then the recognition is multiplied by the number of generation times. The result is the emotion mobilization parameter. The calculation formula is: ; in, A It represents the emotional mobilization parameter; m Indicates the number of likes for comments containing logical relationships; M Indicates the total number of likes for music reviews; N Indicates the number of times it was generated.

[0106] S135 , adding music with the same or similar emotion-mobilizing parameters into a music library corresponding to the emotion-mobilizing parameters to obtain an emotion-mobilizing music library.

[0107] The purpose of this step is that when adjusting the wearer's emotions, it is obviously necessary to be able to establish a recommended music library so that the wearer's emotions can be continuously adjusted. The purpose of this step is to establish a corresponding emotion-mobilizing music library.

[0108] Among them, all the music is analyzed for its comments and the emotional mobilization parameters are obtained.

[0109] Among them, all emotional mobilization must correspond to the corresponding music, so as to establish a corresponding relationship between the two.

[0110] For the obtained same or similar emotion mobilization amounts, the corresponding music is obtained, and these music are all established in the corresponding emotion mobilization music library.

[0111] The obtained emotion-mobilizing music library must also be associated with the emotion-mobilizing parameters corresponding to the music therein, thereby forming a special emotion-mobilizing music library.

[0112] The beneficial effect of step S130 is that, by analyzing the comments of the music, the accuracy of the analysis of the emotions of the analyzed music can be guaranteed.

[0113] As described in step S140, the purpose of this step is to determine the wearer's mood and the emotion adjustment parameters obtained for the music, obtain the wearer's emotion adjustment requirements based on these two parameters, determine the emotion adjustment parameters that need to be applied based on the requirements, and make music recommendations. Specifically: S141: Obtaining a matching degree between the emotion mobilization parameter and the wearer's emotion parameter.

[0114] The purpose of this step is that when making music recommendations, the technical idea of ​​this application is to recommend the music in the emotion-mobilizing music library corresponding to the emotion-mobilizing parameters to the wearer. Therefore, the most basic work is to determine the matching degree between the emotion-mobilizing parameters and the wearer's emotion parameters.

[0115] Among them, the wearer's emotional parameters have been obtained, and they need to be processed to determine the specific relationship between the two types of data to obtain the matching degree between the two.

[0116] After the wearer's emotional parameter is determined, the type of emotion that the wearer needs to adjust or maintain can actually be determined based on the parameter.

[0117] Among them, after obtaining the emotion mobilization parameters, the emotional type of the music itself can also be obtained based on the logical relationship.

[0118] Among them, based on the inherent emotional type of the music and the wearer's emotional type, a corresponding relationship is obtained, so as to select the inherent emotional type of the music that is adapted to the wearer's emotional type. For example: if the wearer is in a depressed stage, the inherent emotional type of the selected music is positive.

[0119] Among them, the process of determining the wearer's emotional parameters and emotional mobilization parameters can be processed based on various suitable methods such as Euclidean distance, weight distribution, etc., and this application does not limit it.

[0120] S142: Sort the matching degrees to obtain a matching degree priority sequence.

[0121] The purpose of this step is that after selecting the emotional type of the music itself, even if the music is similar, its emotional mobilization parameters are obviously not necessarily the same. Among them, the higher the matching degree, the better the emotional adjustment effect on the wearer. Therefore, in the processing, by setting the matching priority sequence, the priority of music selection can be determined.

[0122] Among them, all the obtained matching degrees need to be sorted, and at the same time, the obtained matching degrees need to be associated with the corresponding emotion mobilization parameters.

[0123] S143. Based on the matching degree priority sequence, obtain the emotion-arousing music library that is at the top of the matching degree priority sequence.

[0124] The purpose of this step is to select the corresponding emotion-arousing music library after obtaining the matching priority sequence.

[0125] The so-called top rank of the matching priority sequence may be determined based on the number of music libraries that need to be collected from the selected emotion-arousing music library.

[0126] Here, all the emotion-mobilizing parameters in the front row are obtained, and based on the corresponding relationship between the emotion-mobilizing parameters and the emotion-mobilizing music library, the corresponding emotion-mobilizing music library is determined.

[0127] S144: Based on the ratio of the matching degrees, randomly select music from all the emotion-arousing music libraries according to the ratio, and set the music to the expected recommendation database.

[0128] The purpose of this step is to include all the music recommended to the wearer into the same database. All the music in the database can be recommended to the wearer. In order to reasonably adjust the wearer's emotions, music with the same emotions but different adjustment effects is needed to avoid excessive emotional adjustment for the wearer.

[0129] For the obtained matching degrees, the ratios between the matching degrees are directly calculated, and the ratios of the individual matching degrees and the total matching degrees are obtained.

[0130] The number of music pieces to be selected from different emotion mobilization databases is determined based on the obtained ratios.

[0131] Here, based on the determined number of music, music is randomly selected from the expected recommended music library, and then the selected music is set in the music library. The obtained music library is the expected recommended music library.

[0132] S145. Randomly determine music recommended to the wearer from the expected recommendation database, where the recommended music is the expected recommended music.

[0133] The purpose of this step is to recommend music to the wearer.

[0134] Among them, for the music in the expected recommendation database, random recommendations are made and played to the wearer.

[0135] The beneficial effect of step S140 is that the emotional mobilization parameters of the music and the emotional parameters of the wearer are analyzed to obtain the matching degree between the two. At the same time, the emotional type of the music is selected based on the emotional type of the wearer. Then, the matching degree is calculated according to the emotional type, and an expected recommended music library is established based on the matching degree. The music in the expected recommended music library has different emotional mobilization parameters, and the music with different emotional mobilization parameters for the wearer is adjusted differently to achieve better emotional adjustment effect.

[0136] As described in step S150, the purpose of this step is to ensure that the wearer's mood can change continuously after music is recommended to the wearer. In order to achieve full-process adjustment of the wearer's mood, it is necessary to further adjust the wearer's mood based on the new mood data after the wearer listens to the expected recommended music. Specifically, it includes: S151. Continuously obtain physiological characteristic data of the wearer while listening to the expected recommended music to obtain changed physiological characteristic data.

[0137] The purpose of this step is that when the wearer's mood changes while he or she continues to listen to music, the physiological characteristic data will inevitably change accordingly. Therefore, it is necessary to obtain the physiological characteristic data of the new stage, determine the mood based on the data, and make further adjustments.

[0138] The portable device continuously measures the physiological characteristic data of the wearer during the period of listening to music, and obtains the physiological characteristic data during the period of listening to music, which is the changed physiological characteristic data.

[0139] S152: Obtain the change value of the changed physiological characteristic data to obtain the time starting point of the emotion change.

[0140] The purpose of this step is to take into account that although the wearer's emotions usually change gradually, there will be a point in time when the emotions change significantly, and this significant change will obviously be reflected in the physiological characteristic data. Therefore, the starting point of the emotional change can be determined based on the volatility.

[0141] The fluctuation rate of the changed physiological characteristic data is calculated by obtaining the difference between each type of changed physiological characteristic data and the initial physiological characteristic data, and the result obtained is the change value.

[0142] A threshold is set for the change value. When the change value exceeds the threshold, it is considered that the current wearer's mood has changed, and the time when the mood changed is recorded.

[0143] S153: At the starting point of the emotion change, obtain the changed emotion parameter.

[0144] The purpose of this step is to directly obtain the current emotional parameters when the starting point of the emotional change is determined, and to determine the recommended music based on the emotional parameters.

[0145] Among them, all physiological characteristic data at the starting point of the emotional change are determined, and then the changed emotional parameters are directly calculated based on the physiological characteristic data.

[0146] S154: Obtain the matching degree of the changed emotional parameter and the expected recommendation database corresponding to the expected recommended music to obtain the emotional matching degree of the expected recommended music.

[0147] The purpose of this step is to obtain the changed emotional parameters. In fact, the emotion belongs to the same emotion as the parameters that need to be adjusted at the beginning, but the intensity has changed. In this case, the expected recommended music selected before can actually be applied, but the expected recommended music needs to be further screened.

[0148] Among them, the changed emotional parameters are obtained, and then the matching degree is determined based on the method of step S141, which will not be repeated here.

[0149] The beneficial effect of step S150 is that for new emotion change parameters, there is no need to further search all the music. Instead, based on the matching degree between the expected recommended music and the emotion adjustment, the music type within the new emotion adjustment stage is directly screened out, and the music suitable for recommendation is selected, which reduces the resource consumption in the calculation process.

[0150] As described in step S160, the purpose of this step is that after obtaining new physiological characteristic data, that is, determining new emotional parameters, it is obviously necessary to recommend new music. Therefore, in the specific process, it is necessary to determine the recommended music that can be selected and establish a database. Specifically: S161: Setting a preset emotion matching degree to determine the matching degree between the expected recommended music and the wearer's emotion parameters.

[0151] The purpose of this step is to set a preset emotion matching degree, so as to be used to calculate the obtained specific calculated emotion matching degree, thereby determining whether the screened expected recommended music is applicable.

[0152] Among them, the preset emotion matching degree can be obtained based on the wearer's historical data.

[0153] The preset emotion matching degree can be set according to the matching degree determination method described above, which will not be described in detail here.

[0154] S162. Acquire the physiological characteristic data of the wearer during the listening period of a single piece of expected recommended music, and acquire the emotional parameters during the listening period of a single piece of music.

[0155] The purpose of this step is to more accurately determine the expected recommended music to be used. It is necessary to adjust the mood of each piece of music, so it is necessary to collect the emotional parameters of each piece of music during listening.

[0156] Among them, the emotional parameters of the wearer during listening to the recommended music are collected. The specific calculation method is the same as above and will not be repeated here.

[0157] S163: Obtain a matching degree between the emotion parameter during the listening period of the single piece of music and the emotion mobilization parameter to obtain an emotion matching degree.

[0158] The purpose of this step is to obtain the emotion adjustment parameters of the music when listening to a single piece of music, so as to determine whether the currently recommended music matches the emotion parameters that need to be adjusted. If they match, the music can obviously be used.

[0159] Among them, the matching calculation method for the emotional parameters and emotional mobilization during the listening period of a single piece of music is the same as the matching calculation method mentioned above, which will not be repeated here. The matching result obtained is the emotional matching.

[0160] S164: Obtain music whose emotion matching degree is not lower than a preset emotion matching degree, and input the music into a recommended music database.

[0161] The purpose of this step is to determine whether the corresponding music can be applied based on the obtained emotion matching degree. Therefore, in the specific determination, the result should be determined based on the emotion matching degree.

[0162] The preset emotion matching degree may be determined based on historical data or other methods, which is not limited in this application.

[0163] Among them, the emotion matching degree is compared with the preset emotion matching degree. When the former is not lower than the latter, the music corresponding to the former is considered to be usable music and can be directly entered into the recommended music database.

[0164] S165: establishing an association between the recommended music database and the wearer's emotional parameters during listening to a single piece of music, so as to determine the recommended music database based on the wearer's emotional parameters.

[0165] The purpose of this step is to recommend new music when the wearer's mood changes. In the specific process, in order to obtain more accurate new cycle music recommendation results, it is necessary to establish a recommended music database.

[0166] Among them, the emotional parameters during the listening of a single piece of music are obtained, and the parameters are associated with the recommended music database.

[0167] However, when the wearer listens to music, the duration of the music is usually 3 to 5 minutes, which is very short compared to the total time of the emotion adjustment operation. Therefore, when obtaining the emotion matching degree during the listening period of a single piece of music, it is necessary to determine the acquisition process of the emotion parameters. Specifically: S15 (1) During the recommendation of a single expected music recommendation, setting the evaluation time step of the music listening process.

[0168] The purpose of this step is that the music length of the obtained music listening process is relatively short, so the evaluation time step needs to be re-determined during the processing.

[0169] Among them, for the evaluation time length of the music listening process, the time of the music itself can be evenly segmented, and the time length of each segment is the evaluation time step of the music listening process.

[0170] S15 (2) Based on the evaluation time step of the music listening process, the emotion matching degree during the listening of all single music pieces is obtained.

[0171] The purpose of this step is to obtain the matching degree of each evaluation time step of the music during the music listening period, so as to determine whether the music type meets the wearer's emotional adjustment needs based on the matching degree.

[0172] The matching degree obtained in each evaluation time period of the music listening process is obtained.

[0173] Among them, the emotion matching method is the same as the method mentioned above and will not be repeated here.

[0174] The matching degree within each evaluation time step is obtained, and the mean of the matching degree is calculated. The result obtained is the emotional matching degree during the listening period of a single piece of music.

[0175] S15 (3), obtaining the deviation between the emotion matching degree and the preset emotion matching degree, and constructing a deviation curve.

[0176] The purpose of this step is to determine whether the currently selected expected recommended music is applicable after the wearer determines the matching degree and the deviation amount of the preset matching for different music during the listening period.

[0177] Among them, the deviation of the emotional matching degree of the music that has been listened to in the recommended concert database is recorded, and a deviation curve is constructed, such as Figure 2 As shown, this is a deviation curve of a personalized music recommendation method based on physiological characteristics provided in an embodiment of the present application, wherein the gray straight line is a preset deviation, and the black line is a curve drawn based on the deviation value. The intersection of the two is the starting point for the wearer's emotional change in the next stage.

[0178] S15 (4) Based on the deviation curve, the time point of the wearer's emotion change is obtained, and the time point of the emotion change is the starting point of the wearer's emotion change in the next stage.

[0179] The purpose of this step is to obtain a mathematical model for the deviation curve based on the obtained curve result, and then determine the mood changes in the future period of time based on the mathematical model, so as to predict the time when the mood will change further.

[0180] Among them, when the deviation amount obtained is higher than the preset deviation amount, it is considered that the wearer's mood has further changed. The reason is: for the recommended music database, the music therein can adjust the wearer's mood, and it is considered that the deviation amount is very small. However, when the deviation amount increases, it is obvious that the music does not meet the wearer's mood adjustment needs. The type of music remains unchanged, and it can only be that the wearer's mood has changed.

[0181] Here, based on the curve, the time when the deviation amount is higher than the preset deviation amount can also be predicted, and this time point is the starting point for the wearer's mood change in the next stage.

[0182] S15 (5) , based on the wearer's emotions in the next stage, establishing an expected recommended music library for the next stage.

[0183] The purpose of this step is to build the expected recommended music library for the next stage after determining the starting time of the next stage of emotion, so as to ensure the fluency of the recommended music in the process of emotion adjustment.

[0184] Among them, for the emotional changes during the emotional change period, the physiological characteristic data of the next stage can be calculated directly according to the currently collected physiological characteristic data and the change rate of the physiological characteristic data.

[0185] Among them, the wearer's emotion in the next stage is determined based on the physiological characteristic data of the next stage. The calculation method is the same as the emotion determination method mentioned above and will not be repeated here.

[0186] Among them, after determining the wearer's emotion in the next stage, the expected recommended music library in the next stage is determined based on the emotion. The method used here is the same as the method mentioned above and will not be repeated here.

[0187] S15 (6) When the starting point of the wearer's mood change in the next stage is reached, music in the expected recommendation database for the next stage is recommended to the wearer.

[0188] The purpose of this step is to recommend music to the wearer according to the music adjustment process.

[0189] Among them, the timing system in the interactive device is used to obtain the starting point of the wearer's emotional change in the next stage, and when that time point is reached, the music library is immediately switched to the expected recommendation database for the next stage, and the music in it is recommended.

[0190] Among them, in the process of recommending music in the expected recommendation database of the next stage, the same method is used to determine the expected recommended music library of the next stage, and music recommendation is performed when the corresponding emotion change time point is reached.

[0191] The beneficial effect of step S160 is that it can analyze the mood adjustment effect during music listening, then predict the mood change in the next stage, and based on the prediction results, establish a music recommendation database in advance and make music recommendations in the database.

[0192] As described in step S170, the purpose of this step is to make effective music recommendations. Specifically: S171. Based on the wearer's emotional parameter and the preset emotional matching degree, obtain music that matches the wearer's emotional parameter to obtain available expected recommended music.

[0193] The purpose of this step is to filter other music from the music library that can be used in the mood adjustment process.

[0194] Among them, the wearer's emotional parameters are obtained, and the matching degree between the wearer's emotional parameters and the preset emotions is obtained. Then, based on the matching degree, music that matches the current wearer's emotional parameters is obtained, and the result obtained is the available expected recommended music.

[0195] Among them, in the selection of specific available expected recommended music, based on the matching degree determination method, the emotion mobilization parameters corresponding to the current wearer's emotion parameters are found, and then the music with the same emotion mobilization parameters is obtained to obtain the available expected recommended music.

[0196] In some embodiments, based on the expected recommended music that has been obtained, according to the logical relationship of the emotional words obtained in the sentiment analysis process, based on the logical relationship, available music is screened from the music library to obtain the available expected recommended music.

[0197] S172: Compare the available expected recommended music with the music already in the recommended music database, obtain the available expected recommended music in the non-recommended music library, and add it to the recommended music database.

[0198] The purpose of this step is to further obtain available expected recommended music that can be added to the recommended music database.

[0199] Among them, for all the obtained available expected recommended music, analyze whether they contain available expected recommended music that already exists in the recommended music database. If so, remove it from the available expected recommended music to avoid duplication of music in the recommended music database.

[0200] S173: Randomly sort the music in the recommended music database and recommend it to the wearer.

[0201] The purpose of this step is to perform specific music recommendation operations to the wearer.

[0202] Among them, the music in the recommended music database is randomly sorted, and then the music is recommended to the wearer according to the sorting results.

[0203] The beneficial effect of step S170 is that the music in the recommended music database is expanded and the music therein is randomly sorted. This type of music may contain a variety of styles. Based on the random sorting method, the wearer can feel the diversity of music.

[0204] Those skilled in the art will understand that all or part of the steps of the above-mentioned method embodiments can be implemented by hardware related to computer program instructions, and the aforementioned computer program can be stored in a non-volatile storage medium. When the computer program is executed, it executes the steps of the above-mentioned method embodiments. Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the embodiment of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a non-volatile storage medium and includes a number of instructions for enabling an electronic device (which can be a personal computer, server, network device, etc.) to execute all or part of the methods described in each embodiment of the present invention.

[0205] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. A personalized music recommendation method based on physiological characteristics, characterized in that: The recommended methods include: Based on the portable device, the wearer's physiological characteristic data is obtained and sent to the interactive device; The interactive device acquires the wearer's emotions based on the physiological characteristic data to obtain the wearer's emotional parameters; The interactive device obtains music reviews and obtains emotion mobilization parameters corresponding to the music; providing the wearer with expected recommended music based on the emotion mobilization parameter and the wearer's emotion parameter; Acquiring physiological characteristic data of the wearer when listening to the expected recommended music, and obtaining the emotional matching degree of the expected recommended music; Inputting the music with an emotion matching degree not less than a preset emotion matching degree into a recommended music database, and associating the recommended music database with the wearer's emotion parameter; Based on the wearer's emotional parameters, the available expected recommended music in the non-recommended music database is added to the recommended music database to perform personalized music recommendations.

2. The method for personalized music recommendation based on physiological characteristics according to claim 1, characterized in that: The method of acquiring the wearer's physiological characteristic data based on the portable device and sending the data to the interactive device includes: The portable device continuously monitors physiological characteristics of the wearer and generates physiological characteristic data; A communication relationship is established between the portable device and the interactive equipment, and the physiological characteristic data is sent to the interactive equipment.

3. The method for personalized music recommendation based on physiological characteristics according to claim 1, characterized in that: The interactive device acquires the wearer's emotions based on the physiological characteristic data to obtain the wearer's emotional parameters, including: Acquiring the physiological characteristic data and removing interference items to obtain usable physiological characteristic data; Acquire mutual influences between the available physiological characteristic data, obtain correlation relationships between the available physiological characteristic data, and acquire an available physiological characteristic data group based on the correlation relationships; Selecting one available physiological characteristic data from each of the available physiological characteristic data groups, and obtaining all available physiological characteristic data combinations based on the available physiological characteristic data groups to construct an emotion parameter analysis group; Based on the emotional parameter analysis groups, the wearer's emotional parameters corresponding to each of the emotional parameter analysis groups are obtained.

4. The method for personalized music recommendation based on physiological characteristics according to claim 3, characterized in that: Before determining the wearer's emotional parameter, the method further includes: Setting an evaluation time step for the physiological characteristic data; Based on the evaluation time step, performing a fluctuation measurement on the physiological characteristic data to obtain a fluctuation amplitude; Acquiring physiological characteristic data whose fluctuation amplitude is not less than a preset fluctuation amplitude, and acquiring a time point at which the fluctuation of the physiological characteristic data occurs; Obtaining physiological characteristic data within each evaluation time step after the time node at which the fluctuation occurs, obtaining the fluctuation amplitude of the physiological characteristic data within adjacent evaluation steps, and obtaining adjacent fluctuation amplitudes; Comparing the adjacent fluctuation amplitudes with preset adjacent fluctuation amplitudes, and obtaining a duration in which the adjacent fluctuation amplitudes are not higher than the preset adjacent fluctuation amplitudes; When the duration is not less than the preset duration, it is determined that the wearer's emotional parameter needs to be obtained.

5. The method for personalized music recommendation based on physiological characteristics according to claim 1, characterized in that: The interactive device obtains music reviews and obtains emotion mobilization parameters corresponding to the music, including: Obtain music reviews from the music comment section and obtain all emotional words in the music reviews; Based on the association relationship between all the emotion words, obtaining the logical relationship between all the emotion words; Obtain all the logical relationships in the music review, and obtain the number of times all the logical relationships are generated; Based on the number of times generated, obtaining an emotion mobilization parameter of the music corresponding to the music review; The music with the same or similar emotion-mobilizing parameters is added to the music library corresponding to the emotion-mobilizing parameters to obtain an emotion-mobilizing music library.

6. The method for personalized music recommendation based on physiological characteristics according to claim 1, characterized in that: The providing the wearer with expected recommended music based on the emotion mobilization parameter and the wearer's emotion parameter includes: Obtaining a matching degree between the emotion mobilization parameter and the wearer's emotion parameter; Sorting the matching degrees to obtain a matching degree priority sequence; Based on the matching priority sequence, obtaining the emotion-arousing music library at the top of the matching priority sequence; Based on the matching ratio, randomly selecting music from all the emotion-arousing music libraries according to the ratio, and setting the music into the expected recommendation database; The music recommended to the wearer is randomly determined from the expected recommendation database, and the recommended music is the expected recommended music.

7. The method for personalized music recommendation based on physiological characteristics according to claim 1, characterized in that: The step of obtaining physiological characteristic data of the wearer when listening to the expected recommended music and obtaining the emotional matching degree of the expected recommended music includes: Continuously acquiring physiological characteristic data of the wearer while the wearer listens to the expected recommended music to obtain changed physiological characteristic data; Obtaining a change value of the changed physiological characteristic data to obtain a time starting point of the emotion change; At the starting point of the emotion change, obtaining the changed emotion parameter; The changed emotional parameter and the expected recommendation database matching degree corresponding to the expected recommended music are obtained to obtain the emotional matching degree of the expected recommended music.

8. The method for personalized music recommendation based on physiological characteristics according to claim 1, characterized in that: The step of inputting the music having an emotion matching degree not less than a preset emotion matching degree into a recommended music database, and associating the recommended music database with the wearer's emotion parameter, comprises: Setting a preset emotion matching degree to determine the matching degree between the expected recommended music and the wearer's emotion parameters; Acquiring physiological characteristic data of the wearer during a listening period of a single piece of the expected recommended music, and acquiring emotional parameters during a listening period of a single piece of music; Obtaining a matching degree between the emotion parameter during the listening period of the single piece of music and the emotion mobilization parameter to obtain an emotion matching degree; Acquire music whose emotion matching degree is not less than a preset emotion matching degree, and input the music into a recommended music database; The recommended music database is associated with the wearer's emotional parameters during listening to a single piece of music, so as to determine the recommended music database based on the wearer's emotional parameters.

9. The method for personalized music recommendation based on physiological characteristics according to claim 8, characterized in that: The process of acquiring the emotional parameters during listening to a single piece of music also includes: During a single period of the expected recommended music recommendation, setting an evaluation time step of the music listening process; Obtaining the emotional matching degree of all single music listening periods based on the evaluation time step of the music listening process; Obtaining a deviation between the emotion matching degree and the preset emotion matching degree, and constructing a deviation curve; Based on the deviation curve, obtaining the wearer's mood change time point, the mood change time point being the starting point of the wearer's mood change in the next stage; Obtain the wearer's emotions in the next stage and build the expected recommended music library for the next stage; When the starting point of the wearer's mood change in the next stage is reached, music in the expected recommendation database for the next stage is recommended to the wearer.

10. The method for personalized music recommendation based on physiological characteristics according to claim 1, characterized in that: The method of adding the available expected recommended music in the non-recommended music database to the recommended music database based on the wearer's emotional parameter to perform personalized music recommendation includes: Based on the wearer's emotional parameter and the preset emotional matching degree, obtaining music that matches the wearer's emotional parameter to obtain available expected recommended music; Comparing the available expected recommended music with the music already in the recommended music database, obtaining the available expected recommended music in the non-recommended music library, and adding the music to the recommended music database; The music in the recommended music database is randomly sorted and recommended to the wearer.

Images