CN117093742A - Music data recommendation method and device - Google Patents

Abstract

This application provides a music data recommendation method and device, which can be used in various scenarios such as vehicles. The method includes: obtaining the current navigation destination information of the device and the environmental information of the current location of the device, and obtaining the current navigation destination information. Music data that matches the environmental information; obtains the preferred music type data of the object in the device in the music application of the device; filters the music data according to the preferred music type data to obtain the filtered music data; obtain candidate music data whose similarity or correlation with the filtered music data satisfies preset conditions; generate a recommended music data list based on the filtered music data and the candidate music data , the recommended music data list is used to recommend to objects in the device. Using the solution of this application, the recommended music data can be more in line with the current user's mood and environmental status, and more in line with preferences.

Description

Music data recommendation method and device

Technical field

This application belongs to the technical field of music recommendation, and specifically relates to a music data recommendation method and device.

Background technique

Music applications can recommend music playlists for people who use the music application. However, most of the existing technologies only analyze the historical music data of the people who use the music application in the music application, and recommend based on the data analysis results similar to the historical music data. Music, the music types in the music playlists recommended to users are mostly the same as the music that users have played many times in music applications. It is difficult for users to access other types of music, and users are unknowingly trapped in Information cocoon room, and the existing technology does not make recommendations based on the user's current environment, and cannot meet the user's preference for listening to music in the current environment and in line with the current mood.

Contents of the invention

In view of the objective defects of the aforementioned prior art, in order to solve the above technical problems, the present invention provides a music data recommendation method and device to solve the problem that the recommended music data does not match the user's current mood and the current scene situation, and the music type is excessive. A single question makes recommendations more accurate.

In order to achieve the above objects and other related objects, the present invention provides a music data recommendation method. The method includes: obtaining the current navigation destination information of the device and the environmental information of the current location of the device; obtaining the current navigation destination information. Music data that matches the environmental information; obtains the preferred music type data of the object in the device in the music application of the device; filters the music data according to the preferred music type data to obtain the filtered music data; obtaining candidate music data whose similarity or degree of association with the filtered music data satisfies preset conditions; generating a recommended music data list based on the filtered music data and the candidate music data, The recommended music data list is used to recommend to objects in the device.

In one embodiment of the present invention, obtaining music data matching the current navigation destination information and the environmental information includes: generating at least two pieces of music data based on the current navigation destination information and the environmental information. Genre tag; obtain music data matching the music genre tag from a music database or music streaming platform.

In one embodiment of the present invention, obtaining the preferred music type data of the object in the device in the music application in the device includes: obtaining the different types of music collected by the object in the device in the music application. music data; the different types are at least one of different music style types, different singer types, different language types, and different era types; the first preset number of pieces of music in the different types of music data collected The data is determined to be the preferred music type data.

In one embodiment of the present invention, obtaining the preferred music type data of the object in the device in the music application in the device includes: obtaining the preset time of the object in the device in the music application. Different types of music data played in the segment, the different types are at least one of different music styles, different singers, different languages, and different eras; the different types played in the preset time period The first preset number of pieces of music data in the music data is determined as the preferred music type data.

In one embodiment of the present invention, filtering the music data according to the preferred music type data to obtain filtered music data includes: generating at least two preferred music type data based on the preferred music type data. Tags; determining target music data matching the at least two preferred music type data tags from the music data; using the target music data as filtered music data.

In one embodiment of the present invention, obtaining candidate music data whose similarity or correlation with the filtered music data meets preset conditions includes: deleting the filtered music data from the music data. music data to obtain deleted music data; determine from the deleted music data music data whose similarity or degree of association with the filtered music data satisfies preset conditions to obtain candidate music data, so The preset condition is that the similarity is greater than a preset similarity threshold or the degree of correlation is greater than a preset correlation threshold.

In one embodiment of the present invention, generating a recommended music data list based on the filtered music data and the candidate music data includes: comparing the filtered music data and the candidate music data according to different music types. The music data is sorted to generate a list of candidate recommended music data; when the number of consecutive pieces of music data of the same music type is greater than the preset number, at least one piece of other music is inserted at the target position of the consecutive number of pieces of music data of the same music type. type of music data to obtain the recommended music data list; the target position is located between the first position and the second position, and the first position is the first piece of music data among the consecutive pieces of music data of the same music type. The second position is the position where the preset number of pieces of music data among the consecutive pieces of music data of the same music type is located.

In one embodiment of the present invention, after generating a recommended music data list based on the filtered music data and the candidate music data, the method further includes: sending the music data in the recommended music data list to the object in the device; obtain music feedback information sent by the object in the device based on the music data in the recommended music data list; information based on the information type of the music feedback information and the music feedback information quantity to adjust the music data in the recommended music data list.

In one embodiment of the present invention, the music feedback information includes first feedback information and second feedback information. The first feedback information is used to represent that the object in the device likes the music data or collects the music. data, the second feedback information is used to represent that the object in the device does not like the music data, and the recommended music data list is based on the information type of the music feedback information and the information quantity of the music feedback information. Adjusting the music data in the music data includes: when the music feedback information is the first feedback information and the amount of the music feedback information is greater than a preset amount, adding the first target music data in the recommended music data list, so The first target music data is music data whose similarity or correlation with the music data for which the first feedback information is directed satisfies a preset condition; when the music feedback information is the second feedback information and the When the amount of music feedback information is greater than the preset amount, the second target music data in the recommended music data list is reduced, and the second target music data is the similarity with the music data targeted by the second music feedback information. Or music data whose correlation degree reaches a preset condition, and the preset condition is that the similarity reaches a preset similarity threshold or the correlation degree reaches a preset correlation threshold.

On the other hand, the present invention provides a music data recommendation device. The device includes: an acquisition module, configured to acquire the current navigation destination information and the environment information, and acquire the information related to the current navigation destination information and the environment information. music data of the music type matching the environmental information, obtain the preferred music type data of the object in the device in the music application in the device, and obtain the preferred music type data of the object in the device in the recommended music data list Music feedback information sent by the music data; a filtering module for filtering the music data to obtain the filtered music data; a calculation module for calculating the difference between the deleted music data and the filtered music data. The degree of similarity or association between music data; a sorting module for sorting based on the filtered music data and the candidate music data, and generating a list of recommended music data; an output module for outputting the recommended music data List and send the list of recommended music data to the object in the device.

The music data recommendation method proposed by the present invention uses navigation destination information and environmental information to screen out a batch of music data that conforms to the user's current mood state and current specific scene profile, and can introduce objective factors of navigation destination information and environmental information to determine The range of recommended music data, and then based on the preferred music type data of the object in the device, the amount of the music data is filtered out to match the device object's preferences, which can more accurately recommend music data that matches the user's current mood and environmental state. At the same time, it avoids recommending only a single type of music data, breaks through the information cocoon of recommended music data, and enriches the types of recommended music data.

Description of the drawings

In order to more clearly explain the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description The drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting creative efforts.

Figure 1 is a schematic flow chart 1 of a music data recommendation method provided by the present invention.

Figure 2 is a schematic flow chart 2 of a music data recommendation method provided by the present invention.

Figure 3 is a flow chart 3 of a music data recommendation method provided by the present invention.

Figure 4 is a schematic flow chart 4 of a music data recommendation method provided by the present invention.

Figure 5 is a schematic flow chart 5 of a music data recommendation method provided by the present invention.

Figure 6 is a block diagram of a music data recommendation device provided by the present invention.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

It should be noted that the terms "first", "second", etc. in the description and claims of the embodiments of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. order. It is to be understood that the data so used are interchangeable under appropriate circumstances so that the embodiments of the embodiments of the application described herein can be practiced in sequences other than those illustrated or described herein. Therefore, features defined as "first" and "second" may explicitly or implicitly include one or more of these features. In the description of this embodiment, unless otherwise specified, "plurality" means two or more. Furthermore, the terms "include" and "having" and any variations thereof are intended to cover non-exclusive inclusions, e.g., a process, method, system, product or server that encompasses a series of steps or units and need not be limited to those explicitly listed. Those steps or elements may instead include other steps or elements not expressly listed or inherent to the process, method, product or apparatus.

In order to make the purpose, technical solutions and advantages disclosed in the embodiments of the present application clearer, the embodiments of the present application will be further described in detail below in conjunction with the drawings and examples. It should be understood that the specific embodiments described here are only used to explain the embodiments of the present application and are not used to limit the embodiments of the present application.

Figure 1 is a schematic flowchart 1 of a music data recommendation method provided by the present invention. Figure 2 is a schematic flowchart 2 of a music data recommendation method provided by the present invention. Figure 3 is a schematic flowchart 2 of a music data recommendation method provided by the present invention. Flowchart 3. Figure 4 is a schematic flowchart 4 of a music data recommendation method provided by the present invention. Figure 5 is a schematic flowchart 5 of a music data recommendation method provided by the present invention. Figure 6 is a music data recommendation method provided by the present invention. Block diagram of the data recommendation device.

This specification provides the above-mentioned method operation steps as examples or flow charts, but more or less operation steps may be included based on conventional or non-inventive efforts. The sequence of steps listed in the embodiment is only one way of executing the sequence of many steps, and does not represent the only execution sequence. When the actual system or server product is executed, it may be executed sequentially or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the methods shown in the embodiments or drawings. The present invention provides a music data recommendation method, as shown in Figure 1, which at least includes the following steps:

S102: Obtain the current navigation destination information of the device and the environmental information of the current location of the device;

Optionally, the current navigation destination information of the device is the navigation destination information obtained from the vehicle navigation system.

Optionally, the environmental information may include weather information, geographical information around the navigation destination, the time period during which the vehicle is traveling, traffic conditions at the location, etc. For example, weather information may include weather type, temperature, air quality conditions, etc. For example, the weather type is sunny, cloudy, cloudy, heavy rain, etc., the temperature is 30 degrees Celsius or 25 degrees Celsius, and the air quality conditions are excellent, good, or mild. pollution etc.

Optionally, the device may include a vehicle-mounted intelligent system, a smartphone, a smart computer, etc.

S104: Obtain music data matching the current navigation destination information and the environment information;

Optionally, music type tags can be extracted from the current navigation destination information and the environment information. For example, the music type tags can be "park", "square", "cloudy", "cool", etc.; The music data is music data that matches the tag corresponding to the current navigation purpose information and the environment information. This application does not limit the type of the music data. For example, the music data can be composed of "singer name" Music represented by a combination of tags such as "era" and "style".

S106: Obtain the preferred music type data of the object in the device in the music application in the device;

Optionally, the source of the preferred music type data may be the historical data of the object in the device in the music application in the device. For example, it may include the first preset number of pieces of music played by the object in the device within the preset time period in the music application, different types of music data collected in the music application, and the preset time period in the music application. The first preset number of music, etc., sorted from high to low by playback volume.

The object in the device refers to the account of the registered user in the device. Optionally, the account of the registered user in the device can be the account registered in the vehicle-mounted intelligent system by the user operating the vehicle, or the account registered by the user of the smartphone in the smart phone. The account registered in the music app on your phone.

S108: Filter the music data according to the preferred music type data to obtain filtered music data;

Optionally, two or more preferred music type data tags are generated based on the preferred music type, and target music data that simultaneously matches the at least two preferred music type data tags is determined from the music data. This application does not limit preferences. The type of music type data. For example, the preferred music type data is obtained from the historical data of the object in the device in the music application, and can be represented by a combination of tags such as "singer name", "age", and "style". music data.

S110: Obtain candidate music data whose similarity or correlation with the filtered music data satisfies preset conditions;

Optionally, meeting the preset condition means that the similarity reaches the preset similarity threshold or the degree of correlation reaches the preset correlation threshold. Among the music data that conforms to the current user's mood and environmental information, recalling music data that is more in line with the subject's preference needs, that is, is similar or related, but does not completely match the subject's preferences, can enrich the recommended music data. type.

The similarity refers to the degree of similarity between the features of the music data, and the degree of correlation refers to the correlation between the features of the music data. The preset condition is that the similarity is greater than the preset similarity threshold or the degree of association is greater than the preset Set the correlation threshold.

S112: Based on the filtered music data and the candidate music data, generate a recommended music data list, where the recommended music data list is used to recommend to objects in the device.

Optionally, the filtered music data and candidate music data can be sorted in a certain manner, and a recommended music data list is generated after sorting, for example, sorting using diversity enhancement.

The music data recommendation method proposed by the present invention uses navigation destination information and environmental information to screen out a batch of music data that conforms to the user's current mood state and specific scene profile, and can introduce objective factors of navigation destination information and environmental information to determine recommendations. The range of music data, and then based on the preferred music type data of the object in the device, the amount of music data is filtered out to match the preferences of the device object, which can more accurately recommend music data that matches the user's current mood and environmental state, breaking through It not only provides information cocoon for recommended music data, but also enriches the types of recommended music data.

In one embodiment of the present invention, obtaining music data matching the current navigation destination information and the environmental information includes: generating at least two pieces of music data based on the current navigation destination information and the environmental information. Genre tag; obtain music data matching the music genre tag from a music database or music streaming platform.

This embodiment can break through the information cocoon of music data recommendation by acquiring current navigation destination information and environmental information, and selecting music data determined by external environmental factors based on these two external environmental factors.

This application does not specifically limit the music type tag. For example, the music type tag refers to the tag of the music type that matches the music data, for example, "country", "lyrical", "Cantonese", and the music database or streaming media platform refers to It is a platform or database that stores music data and can be played. For example, if the current navigation destination information is "a certain suit customization shop" and the weather type is sunny, the music type tags can be generated as "suit" and "sunny", and the two music types "suit" and "sunny" can be obtained from the music database. Music data to which the genre tag matches.

As mentioned above, the source of preferred music type data may include the first preset number of pieces of music played by the object in the device within a preset time period in the music application, different types of music data collected in the music application, The first preset number of songs sorted from high to low by playback volume within the preset time period, etc. In another embodiment provided by the present invention, specifically, Figure 2 provides a schematic flowchart 2 of a music data recommendation method. As shown in Figure 2, the above step 106 may also include:

S1061: Obtain different types of music data collected by the object in the device in the music application, and determine the first preset number of pieces of music data among the different types of music data collected as preferred music type data.

This application does not specifically limit the type of music. For example, it can be "rock", "cowboy" and "country". The preset number can be set according to actual needs. There is no specific limit here. For example, the preset number can be set according to actual needs. If the number is 50 or 100, then step S106 is to obtain the preferred music type data of the first 50 pieces of music or the first 100 pieces of music collected by the object in the device in the music application.

In another embodiment provided by the present invention, specifically, Figure 3 provides a schematic flowchart 3 of a music data recommendation method. As shown in Figure 3, the above step 106 may also include:

S1062: Obtain different types of music data played by objects in the device within a preset time period in the music application, and add the previously preset number of different types of music data played within the preset time period. The piece of music data is determined to be the preferred music type data.

The preset time period can be set according to actual needs, and is not specifically limited here. For example, it can be within the first 30 days from the date of obtaining the music data.

The different types in S1061 and S1062 are at least one of different types such as different music style types, different singer types, different language types, different era types, etc. For example, the music style type can be pop style, country style, rock style, the singer type can be Chinese singer, Japanese and Korean singer, etc., and the language type can be Chinese, Cantonese, Russian, English, etc.

In addition, the source of the preferred music type data obtained in step S106 can also be the first preset number of pieces of music that are sorted from high to low by play volume within the preset time period in the music application by the object in the device, for example, Preferred music type data of the subject's top 50 or top 100 pieces of music sorted by play volume within the preset time period in the music application.

This embodiment of the present invention sets up multiple sources that can obtain preferred music type data, can take into account the usage of different objects in music applications, use different sources to obtain the object's preferred music type data, and can better evaluate the object's interest in music. Preferences.

In one embodiment of the present invention, filtering the music data according to the preferred music type data to obtain filtered music data includes:

Generate at least two preferred music type data tags based on the preferred music type data; determine target music data matching the at least two preferred music type data tags from the music data; use the target music data as Filtered music data.

This embodiment of the present invention can determine the target music data taking into account the subject's preference for music data, and is more in line with the subject's preference needs.

In one embodiment of the present invention, the acquisition of candidate music data whose similarity or correlation with the filtered music data meets preset conditions includes:

Delete the filtered music data from the music data to obtain deleted music data; determine the degree of similarity or correlation between the deleted music data and the filtered music data to satisfy a preset Conditional music data to obtain candidate music data.

The similarity refers to the degree of similarity between the features of the music data, and the degree of correlation refers to the correlation between the features of the music data. The preset condition is that the similarity is greater than the preset similarity threshold or the degree of association is greater than the preset Set the correlation threshold.

The embodiment provided by the present invention can recall music data that is similar or related to the filtered music data for the deleted music data, thereby preventing the music data from being completely determined by the object's preferences.

In one embodiment provided by the present invention, Figure 4 is a flow chart 4 of a music data recommendation method provided by the present invention. As shown in Figure 4, the method based on the filtered music data and the candidate music data , generate a recommended music data list, also including:

S402: Sort the filtered music data and the candidate music data according to different music types, and generate a list of candidate recommended music data;

This application does not limit the music type. For example, different music types can be "country", "rock", "lyrical", etc. The candidate recommended music data list is a music data list generated by first sorting the filtered music data and the candidate music data, and is arranged according to different music types. For example, in the candidate recommended music data list, There are 8 consecutively arranged pieces of music data with the music type "lyrical".

S404: When the number of consecutive pieces of music data of the same music type is greater than the preset number, insert at least one piece of music data of another music type at the target position of the consecutive pieces of music data of the same music type to obtain the recommendation. Music data list; the target position is located between the first position and the second position, the first position is the position of the first piece of music data in the continuous number of music data of the same music type, and the second position is the position of the first piece of music data of the same music type. The location of the preset number of pieces of music data among the consecutive pieces of music data of the same music type.

Optionally, this embodiment uses diversity-enhanced sorting to re-sort the candidate recommended music data list. When the number of music data of the same music type exceeds the preset number, other types of music data are inserted to disrupt the candidate recommendations. For the sorting of the music data list, for example, the preset number is set to 6. When there are 8 consecutive pieces of music data in the candidate recommended music list whose music type is "Jazz", then the 8 pieces of music type are "Jazz". Insert other types of music data between the first to sixth pieces of music data of "Jazz", such as inserting a piece of music data with a music style of "lyrical".

After the candidate recommended music data list is scrambled, the recommended music data list is obtained, in which the number of music data of the same music type for each consecutive number does not exceed the preset number. This application will not elaborate on the preset number. , the preset number can be set according to the actual situation. For example, if the preset number is set to 4, then there cannot be music data of 5 consecutive pieces by the same singer, nor can there be 5 consecutive pieces of music of the same language. According to this, when the 5 consecutive Chinese songs are music data of the Chinese language, other types of music are inserted between the position of the first music data of the 5 consecutive Chinese songs and the position of the 4th music data. data; for another example, if the preset number is set to 5, then the number of music data of the same music type for each consecutive number of songs in the recommended music data list can be 2, 3, 4, or 5. There is no consecutive number of songs. The number of music data of the same music type is more than 6 pieces.

The embodiment provided by the present invention can increase the diversity of recommendation results on the basis of maintaining personalized recommendations to avoid over-recommendation of similar songs, so as to increase the user's selectivity and listening experience, and can make recommendations based on the feedback of the object. The music data list is optimized to be more in line with the object's preferences, enhance the randomness of the final generated music data list, reduce the repetition of the same type of music in the recommended music data list, and prevent users from getting bored.

In one embodiment of the present invention, after generating a recommended music data list based on the filtered music data and the candidate music data, the method further includes:

Send music data in the recommended music data list to an object in the device; obtain music feedback information sent by the object in the device based on the music data in the recommended music data list; based on the music feedback information The information type and the information quantity of the music feedback information are used to adjust the music data in the recommended music data list.

Figure 5 is a schematic flow chart 5 of a music data recommendation method provided by the present invention. As shown in Figure 5, in one embodiment of the present invention, the information type based on the music feedback information and the information of the music feedback information Quantity, adjusting the music data in the recommended music data list, also includes:

S502: Obtain music feedback information sent by the object in the device based on the music data in the recommended music data list;

Obtaining music feedback information can be used to evaluate the accuracy of the recommended music data, thereby updating the recommended music data list and optimizing the recommended music data. The music feedback information is used to characterize the likes and dislikes of the objects in the device towards the music data. degree of information.

S504: When the music feedback information is the first feedback information and the quantity of the music feedback information is greater than the preset quantity, add the first target music data in the recommended music data list; when the music feedback information is the When the second feedback information is received and the amount of the music feedback information is greater than the preset amount, the second target music data in the recommended music data list is reduced.

The first feedback information is used to represent that the object in the device likes the music data or collects the music data. If the object in the device collects the recommended music data, the first feedback information is formed, and the second feedback is The information is used to indicate that the object in the device does not like the music data. If the object in the device performs operations such as canceling favorites, switching songs, etc. on the recommended music data, the second feedback information is formed.

When the amount of first feedback information or the amount of second feedback information is greater than the preset amount, the first target music data is added or the second target music data is reduced, and the recommended music data list is updated. The first target music data is the same as the first feedback information. The similarity or correlation between the music data targeted by the information meets the preset conditions, and the second target music data is the music data whose similarity or correlation with the music data targeted by the second music feedback information meets the preset conditions. Conditional music data.

Optionally, in this application, a variety of methods can be used to calculate the similarity and degree of correlation, which are not specifically limited. For example, correlation factors can be used to calculate the similarity and degree of correlation between music data. Satisfying the preset conditions means that the similarity reaches the preset similarity threshold or the degree of correlation reaches the preset correlation threshold. For another example, cosine similarity can be used to calculate the similarity between music data, and different music data can be classified according to feature items and weights. Multiple vectors are formed, and the cosine value of the angle between the two vectors is calculated to evaluate their similarity. The smaller the angle is, it is closer to 0 degrees. The closer the cosine value is to 1, the more consistent their directions are, and the more similar they are; when The cosine of the angle between two vectors with completely opposite directions takes the minimum value -1.

Specifically, if the preset number is set to 20 pieces, and when the number of first feedback information is greater than 20 pieces, the similarity or correlation between each piece of music data corresponding to the number of first feedback information will satisfy the preset condition. Music data is used as the first target music data to reduce recommendations for the first target music data.

Preferably, when the number of at least one music type data tag in the preset number reaches the preset threshold, then the similarity or correlation between each piece of music data corresponding to the first number of feedback information satisfies the preset condition. Music data is used as the first target music data to reduce the recommendation of the first target music data. For example, the default number is set to 20. When the number of first feedback information is greater than 20, 10 of them are related to "Cantonese" and "Nostalgia" Music data that matches the music type data tag of "Love" will be music data whose similarity or correlation with the music data with the music type data tag of "Cantonese", "nostalgia" and "love" meets the preset conditions. The data is used as the first target music data to reduce recommendations for the first target music data.

The embodiment provided by the present invention can classify and identify feedback information, increase or decrease the recommended number of target music data in the list according to the subject's acceptance of music data, more in line with the subject's preferences, and set a preset number so that the feedback information The reflected object's acceptance of music data is more accurate and has stronger referentiality, enabling more personalized and accurate music recommendations.

In addition to the technical field of music recommendation, the present invention can also be applied to other technical fields. After obtaining tags corresponding to navigation destinations and environmental information, in some embodiments, food types matching the tags can be obtained to form food type tags. Finally, by matching the food types of the restaurants in the food APP, and at the same time obtaining the historical data of the food types consumed by the objects in the device in the restaurants, recommended food radio programs, nearby restaurants, for example, the weather in the current environmental information If the type is "Xiao Xue", it matches food types such as "hot pot" and "barbecue", and can recommend playing related food radio programs, or recommend hot pot restaurants, barbecue restaurants, etc. near the navigation destination; in other embodiments, it can recommend Suitable outdoor or indoor activities, for example, the weather type in the current environmental information is "sunny" and the temperature is "28 degrees Celsius", nearby parks or outdoor sports venues can be recommended; in other embodiments, the navigation destination can be Tags corresponding to the environmental information can be used to obtain matching tourist attractions or scenic spots. For example, in winter snowy days, nearby ski resorts or hot spring resorts can be recommended; in other embodiments, labels corresponding to navigation destinations and environmental information can be obtained. Store promotions that match the tags. For example, on a rainy day, the system can recommend umbrella promotions in nearby stores; in other embodiments, the system can recommend suitable health and sports activities based on tags corresponding to navigation destinations and environmental information. .

An embodiment of the present invention also provides a music data recommendation device for implementing the above method and each step. Figure 6 is a block diagram of a music data recommendation device provided by the present invention. As shown in Figure 6, the device may include :

Acquisition module 602, configured to acquire the current navigation destination information and the environment information, acquire music data of a music type that matches the current navigation destination information and the environment information, and acquire objects in the device Prefer music type data in the music application in the device, and obtain music feedback information sent by the object in the device for the music data in the recommended music data list;

A filtering module 604 is used to filter the music data to obtain the filtered music data;

Calculation module 606, used to calculate the degree of similarity or correlation between the deleted music data and the filtered music data;

A sorting module 608 is configured to sort based on the filtered music data and the candidate music data, and generate a recommended music data list.

In an optional embodiment, the acquisition module 602 includes:

A music type tag generating unit, configured to generate at least two music type tags based on the current navigation destination information and the environmental information;

A music data acquisition unit is used to acquire music data matching the music type tag from a music database or a music streaming platform.

In another optional embodiment, the acquisition module 602 also includes:

The first unit for obtaining different types of music data is used to obtain different types of music data played by objects in the device within a preset time period in the music application. The different types are different music styles, different singing At least one of author type, different language type, and different era type;

The first unit for obtaining preferred music type data is used to determine the first preset number of pieces of music data among the different types of music data played within the preset time period as the preferred music type data.

In another optional embodiment, the acquisition module 602 also includes:

The second unit for obtaining different music type data is used to obtain different types of music data played by objects in the device within a preset time period in the music application. The different types are different music styles, different singing At least one of author type, different language type, and different era type;

The second unit for obtaining preferred music type data is used to determine the first preset number of pieces of music data among the different types of music data played within the preset time period as the preferred music type data.

In another optional embodiment, the filtering module 604 includes:

a preferred music type data label generating unit, configured to generate at least two preferred music type data labels based on the preferred music type data;

A target music data determining unit configured to determine, from the music data, target music data that matches the at least two preferred music type data tags;

A filtered music data determination unit is configured to use the target music data as filtered music data.

In another optional embodiment, the acquisition module 602 also includes:

a deleted music data acquisition unit, configured to delete the filtered music data from the music data to obtain deleted music data;

The candidate music data acquisition unit determines, from the deleted music data, music data whose similarity or correlation with the filtered music data meets a preset condition, and obtains candidate music data.

In another optional embodiment, the calculation module 606 includes:

The first similarity or correlation calculation unit is used to calculate the similarity or correlation between the deleted music data and the filtered music data; determine whether the similarity or correlation is greater than the preset similarity threshold or the correlation is greater than Default correlation threshold.

In another optional embodiment, the sorting module 608 includes:

A candidate recommended music data list generating unit, configured to sort the filtered music data and the candidate music data according to different music types, and generate a candidate recommended music data list;

The recommended music data list generating unit is configured to insert at least one piece of other music types at the target position of the consecutive pieces of music data of the same music type when the number of consecutive pieces of music data of the same music type is greater than the preset number. music data.

In another optional embodiment, the acquisition module 602 also includes:

A music feedback information obtaining unit is configured to obtain music feedback information sent by an object in the device based on the music data in the recommended music data list.

In another optional embodiment, the calculation module 606 also includes:

A music feedback information quantity calculation unit configured to calculate the number of first music feedback information and second music feedback information in the music feedback information sent by the object in the device based on the music data in the recommended music data list;

A music feedback information quantity judgment unit is used to determine whether the music feedback information quantity is greater than a preset quantity;

A second similarity or correlation calculation unit is used to calculate the similarity or correlation between the music data targeted by the first feedback information and other music data, and to calculate the similarity or correlation between the music data targeted by the second feedback information and other music data. degree of similarity or relatedness.

In another optional embodiment, the sorting module 608 also includes:

Music data list adjustment module: used to adjust the music data in the recommended music data list based on the information type of the music feedback information and the information quantity of the music feedback information, and add the first music data in the recommended music data list. target music data, or reduce the second target music data in the recommended music data list.

It should be noted that the device embodiments provided in the embodiments of this application are based on the same inventive concept as the above method embodiments.

The music data recommendation method and device provided by the embodiments of this application have the following technical effects:

1. The music data recommendation method proposed by the present invention uses navigation destination information and environmental information to screen out a batch of music data that conforms to the user's current mood state and specific scene profile, and can introduce objective factors of navigation destination information and environmental information to Determine the scope of recommended music data, and then filter the music that matches the preferences of the device object according to the number of music data based on the preferred music type data of the object in the device, which can more accurately recommend music data that matches the user's current mood and environmental state. , breaking through the information cocoon of recommended music data, and at the same time enriching the types of recommended music data;

2. The embodiment provided by the present invention breaks through the information cocoon of music data recommendations by acquiring current navigation destination information and environmental information, and selecting music data determined by external environmental factors based on these two external environmental factors;

3. The embodiment provided by the present invention provides multiple sources for obtaining preferred music type data, can take into account the usage of different objects in music applications, obtain the object's preferred music type data using different methods, and can better evaluate the object. musical preference;

4. The embodiment provided by the present invention can determine the target music data taking into account the subject's preference for music data, and is more in line with the subject's preference needs;

5. The embodiments provided by the present invention can recall music data that is similar or related to the filtered music data for the deleted music data, preventing the music data from being completely determined by the object's preferences;

6. The embodiment provided by the present invention can increase the diversity of recommendation results on the basis of maintaining personalized recommendations to avoid over-recommendation of similar songs, so as to increase the user's selectivity and listening experience, and can be based on the feedback of the object. Optimize the recommended music data list to be more in line with the object's preferences, enhance the randomness of the final generated music data list, reduce the repetition of the same type of music in the recommended music data list, and avoid users getting bored;

7. The embodiments provided by the present invention can classify and identify feedback information, increase or decrease the recommended number of target music data in the list according to the subject's acceptance of the music data, which is more in line with the subject's preferences, and set a preset number to enable feedback The object's acceptance of music data reflected by the information is more accurate and has stronger referentiality, enabling more personalized and accurate music recommendations.

It should be noted that the above-mentioned order of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. Specific embodiments of this specification have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desired results. Additionally, the processes depicted in the figures do not necessarily require the specific order shown, or sequential order, to achieve desirable results. Multitasking and parallel processing are also possible or may be advantageous in certain implementations.

Each embodiment in this specification is described in a progressive manner. The same and similar parts between the various embodiments can be referred to each other. Each embodiment focuses on its differences from other embodiments. In particular, for the device and server embodiments, since they are basically similar to the method embodiments, the descriptions are relatively simple. For relevant details, please refer to the partial description of the method embodiments.

The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application. Inside.

Claims (10)

1. A music data recommendation method, the method comprising:

acquiring current navigation destination information of equipment and environment information of the current position of the equipment;

acquiring music data matched with the current navigation destination information and the environment information;

acquiring preference music type data of an object in a device in a music application in the device;

filtering the music data according to the preference music type data to obtain filtered music data;

acquiring candidate music data with the similarity or association degree between the candidate music data and the filtered music data meeting a preset condition;

and generating a recommended music data list based on the filtered music data and the candidate music data, wherein the recommended music data list is used for recommending the objects in the equipment.

2. The method of claim 1, wherein the acquiring music data that matches the current navigation destination information and the environment information comprises:

generating at least two music type tags based on the current navigation destination information and the environment information;

and acquiring the music data matched with the music type tag from a music database or a music streaming media platform.

3. The method of claim 1, wherein the obtaining of the preference music type data for the object in the device in the music application in the device comprises:

acquiring different types of music data collected by an object in the equipment in the music application; the different types are at least one of different music style types, different singer types, different language types and different age types;

and determining the front preset quantity of pieces of music data in the different types of music data collected as preference music type data.

4. The method of claim 1, wherein the obtaining of the preference music type data for the object in the device in the music application in the device comprises:

acquiring different types of music data played by an object in the equipment in a preset time period in the music application, wherein the different types are at least one of different music style types, different singer types, different language types and different age types;

and determining the previous preset number of pieces of music data in different types of music data played in the preset time period as preference music type data.

5. The method of claim 1, wherein filtering the music data according to the preferred music genre data to obtain filtered music data comprises:

generating at least two preference music type data tags based on the preference music type data;

determining target music data matched with the at least two preference music type data tags from the music data;

and taking the target music data as the filtered music data.

6. The method according to claim 1, wherein the obtaining candidate music data whose degree of similarity or association with the filtered music data satisfies a preset condition includes:

deleting the filtered music data from the music data to obtain deleted music data;

and determining music data with similarity or association degree meeting preset conditions between the deleted music data and the filtered music data to obtain candidate music data, wherein the preset conditions are that the similarity is larger than a preset similarity threshold value or the association degree is larger than a preset association degree threshold value.

7. The method of claim 1, wherein the generating a list of recommended music data based on the filtered music data and the candidate music data comprises:

Sorting the filtered music data and the candidate music data according to different music types to generate a candidate recommended music data list;

when the number of the continuous number of the music data of the same music type is larger than the preset number of the music data, inserting at least one piece of music data of other music types at the target position of the continuous number of the music data of the same music type to obtain the recommended music data list; the target position is located between a first position and a second position, the first position is a position where a first piece of music data in the continuous quantity of pieces of music data with the same music type is located, and the second position is a position where a first preset quantity of pieces of music data in the continuous quantity of pieces of music data with the same music type is located.

8. The method according to any one of claims 1 to 7, wherein after the generating a recommended music data list based on the filtered music data and the candidate music data, the method further comprises:

transmitting music data in the recommended music data list to an object in the device;

acquiring music feedback information sent by an object in the equipment based on the music data in the recommended music data list;

And adjusting the music data in the recommended music data list based on the information type of the music feedback information and the information quantity of the music feedback information.

9. The method of claim 8, wherein the music feedback information includes first feedback information for characterizing that the object in the device likes the music data or favorites the music data, and second feedback information for characterizing that the object in the device dislikes the music data, wherein the adjusting the music data in the recommended music data list based on the information type of the music feedback information and the information amount of the music feedback information includes:

when the music feedback information is the first feedback information and the number of the music feedback information is larger than a preset number, adding first target music data in a recommended music data list, wherein the first target music data are music data with similarity or association degree between the first target music data and the music data aimed at by the first feedback information meeting preset conditions;

and when the music feedback information is the second feedback information and the number of the music feedback information is larger than the preset number, reducing second target music data in a recommended music data list, wherein the second target music data is music data with similarity or association degree between the second target music data and the music data aimed at by the second music feedback information reaching a preset condition, and the preset condition is that the similarity reaches a preset similarity threshold or the association degree reaches a preset association degree threshold.

10. A music data recommendation apparatus, characterized in that the apparatus comprises:

the acquisition module is used for acquiring the current navigation destination information and the environment information, acquiring music data of a music type matched with the current navigation destination information and the environment information, acquiring preference music type data of an object in the equipment in a music application in the equipment, and acquiring music feedback information sent by the object in the equipment aiming at the music data in the recommended music data list;

the filtering module is used for filtering the music data to obtain the filtered music data;

a calculation module, configured to calculate a degree of similarity or association between the deleted music data and the filtered music data;

and the sorting module is used for sorting based on the filtered music data and the candidate music data and generating a recommended music data list.