Information prompt method and device, storage medium and electronic device

Abstract

Embodiments of the present invention relate to an information prompting method, an information prompting device, a storage medium, and an electronic device, and relate to the field of information processing technology. The method comprises: obtaining multimedia objects from a multimedia object list, extracting features related to the multimedia objects; determining a target multimedia object based on the features related to the multimedia objects; extracting features related to segments of the target multimedia objects; determining a target segment based on the features related to the segments; and providing an information prompt for the target segment. The present invention can enhance the conversion effect of information prompts.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of information processing technology. More specifically, embodiments of the present invention relate to an information prompting method, an information prompting device, a computer-readable storage medium, and an electronic device. Background Art

[0002] This section is intended to provide a background or context to embodiments of the invention that are recited in the claims, and no admission is made that the description herein is prior art by inclusion in this section.

[0003] Typically, when a user is playing audio or video, the platform will send some prompt information to the user to prompt the user to perform relevant operations, such as payment conversion information.

[0004] However, in the current process of pushing paid conversion information, it is usually pushed at a fixed time, resulting in poor user experience and poor paid conversion effect. Summary of the Invention

[0005] In this context, embodiments of the present invention are intended to provide an information prompting method, an information prompting device, a computer-readable storage medium, and an electronic device.

[0006] According to a first aspect of an embodiment of the present invention, an information prompting method is provided, comprising: obtaining multimedia objects in a multimedia object list, extracting multimedia object-related features; determining a target multimedia object based on the multimedia object-related features; extracting segment-related features of the target multimedia object; determining a target segment based on the segment-related features; and providing information prompting in the target segment.

[0007] In some embodiments of the present invention, determining the target multimedia object based on the multimedia object-related features includes: determining the single feature score corresponding to each of the first preset number of multimedia objects in the multimedia object list based on the calculation logic of each multimedia object-related feature; then determining the total score of each multimedia object based on the single feature score and the corresponding weight coefficient of each multimedia object-related feature; and determining the multimedia object with the highest total score as the target multimedia object.

[0008] In some embodiments of the present invention, the segment-related features include: playback duration of each dimension and the number of segment type switching times of each dimension.

[0009] In some embodiments of the present invention, determining the target segment based on the segment-related characteristics includes: determining the user's maximum playback time based on the playback time of each dimension; and determining the target segment within the user's maximum playback time range in the target multimedia object based on the number of segment type switches in each dimension.

[0010] In some embodiments of the present invention, determining the maximum playback time of the user based on the playback time of each dimension includes: determining the maximum playback time of the user based on the user playback time, the playback time within the site, and the playback time of the multimedia object, combined with the corresponding weights.

[0011] In some embodiments of the present invention, determining the target segment within the maximum playback time of the user in the target multimedia object based on the number of segment type switches in each dimension includes: determining the segment type with the highest probability of being cut away within the maximum playback time of the user in the target multimedia object based on the number of segment type switches in the multimedia object dimension, the number of segment type switches in the site dimension, and the number of segment type switches in the style dimension, in combination with corresponding weights; and determining the segment corresponding to the segment type with the highest probability of being cut away as the target segment.

[0012] In some embodiments of the present invention, when there are multiple segments corresponding to the maximum probability cut-away segment type; determining the segment corresponding to the maximum probability cut-away segment type as the target segment includes: determining the similarity between the label corresponding to each segment corresponding to the maximum probability cut-away segment type and the user playback label; and determining the segment with the greatest similarity as the target segment.

[0013] According to a second aspect of an embodiment of the present invention, an information prompting device is provided, comprising: a first feature extraction module, configured to extract multimedia object-related features when acquiring multimedia objects in a multimedia object list; a target multimedia object determination module, configured to determine a target multimedia object based on the multimedia object-related features; a second feature extraction module, configured to extract segment-related features of the target multimedia object; a target segment determination module, configured to determine a target segment based on the segment-related features; and an information prompting module, configured to provide information prompts for the target segment.

[0014] In some embodiments of the present invention, the target multimedia object determination module is used to determine the single feature score corresponding to each of the first preset number of multimedia objects in the multimedia object list based on the calculation logic of the relevant features of each multimedia object; then determine the total score of each multimedia object based on the single feature score and the corresponding weight coefficient of each of the relevant features of the multimedia object; and determine the multimedia object with the highest total score as the target multimedia object.

[0015] In some embodiments of the present invention, the segment-related features include: playback duration of each dimension and the number of segment type switching times of each dimension.

[0016] In some embodiments of the present invention, the target segment determination module is used to determine the maximum playback time of the user based on the playback time of each dimension; and determine the target segment within the maximum playback time range of the user in the target multimedia object based on the number of segment type switches in each dimension.

[0017] In some embodiments of the present invention, the target segment determination module is used to determine the maximum playback time of the user based on the user playback time, the site playback time and the multimedia object playback time, combined with corresponding weights.

[0018] In some embodiments of the present invention, the target segment determination module is used to determine the segment type with the highest probability of being cut away within the maximum playback time of the user in the target multimedia object based on the number of segment type switches in the multimedia object dimension, the number of segment type switches in the site dimension, and the number of segment type switches in the style dimension, combined with corresponding weights; and determine the segment corresponding to the segment type with the highest probability of being cut away as the target segment.

[0019] In some embodiments of the present invention, when there are multiple segments corresponding to the maximum probability cut-away segment type; the target segment determination module is used to determine the similarity between the label corresponding to each segment corresponding to the maximum probability cut-away segment type and the user playback label; and determine the segment with the greatest similarity as the target segment.

[0020] In some embodiments of the present invention, the target multimedia object determination module is configured to adjust the weight coefficient of each multimedia object-related feature according to user historical conversion data.

[0021] In some embodiments of the present invention, the multimedia object related features include: one or more of: playback entry, average number of songs listened to, song tags, user song tags, user registration time, and user platform level.

[0022] According to a third aspect of the embodiments of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any one of the above methods is implemented.

[0023] According to a fourth aspect of an embodiment of the present invention, there is provided an electronic device, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform any one of the above methods by executing the executable instructions.

[0024] According to the method, device, computer-readable storage medium and electronic device of the embodiment of the present invention, by obtaining multimedia objects in the multimedia object list and extracting multimedia object-related features, the target multimedia object for information prompting can be determined based on the multimedia object-related features, and then segment-related features are further extracted from the target multimedia object, and the target segment is determined based on the segment-related features to provide information prompts in the target segment, so that information prompts can be provided at a more appropriate time according to different users, thereby improving user experience and increasing the conversion effect of information prompts. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily apparent by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which:

[0026] Figure 1 A flowchart of an information prompting method according to an embodiment of the present invention is shown;

[0027] Figure 2 A structural block diagram of an information prompting device according to an embodiment of the present invention is shown;

[0028] Figure 3 A schematic diagram of a storage medium according to an embodiment of the present invention is shown;

[0029] Figure 4 A structural block diagram of an electronic device according to an embodiment of the present invention is shown.

[0030] In the drawings, the same or corresponding reference numerals denote the same or corresponding parts. DETAILED DESCRIPTION

[0031] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. Rather, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0032] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present invention may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software.

[0033] According to an embodiment of the present invention, an information prompting method, an information prompting device, a computer-readable storage medium, and an electronic device are provided.

[0034] In this document, the numbers of any elements in the drawings are for illustrative purposes only and are not intended to be limiting. Any nomenclature is for distinction only and does not have any limiting meaning. The data involved in this disclosure may be authorized by the user or fully authorized by all parties. The collection, dissemination, and use of this data shall comply with relevant national laws and regulations. The embodiments / examples of this disclosure may be combined with each other.

[0035] The principles and spirit of the present invention are described in detail below with reference to several representative embodiments of the present invention. SUMMARY OF THE INVENTION

[0037] The inventors have discovered that in the current process of pushing paid conversion information, the information is usually pushed at a fixed time, resulting in a poor user experience and poor paid conversion effect.

[0038] In view of the above, the basic idea of ​​the present invention is to provide an information prompt method, an information prompt device, a computer-readable storage medium and an electronic device to push information according to the actual situation of each user, thereby improving user experience and increasing conversion effect.

[0039] After introducing the basic principles of the present invention, various non-limiting embodiments of the present invention are described in detail below.

[0040] Application Scenario Overview

[0041] It should be noted that the following application scenarios are only provided to facilitate understanding of the spirit and principles of the present invention, and the embodiments of the present invention are not limited in this respect. On the contrary, the embodiments of the present invention can be applied to any applicable scenario.

[0042] The present invention can be applied to all scenarios where multimedia objects are applicable, such as audio playback scenarios, video playback scenarios, or electronic document reading scenarios.

[0043] Exemplary Methods

[0044] Reference below Figure 1 The information prompting method according to an exemplary embodiment of the present invention will be described.

[0045] Figure 1 Schematically shows a flow chart of an information prompting method according to an exemplary embodiment of the present invention. Figure 1 According to an exemplary embodiment of the present invention, the information prompting method may include the following steps:

[0046] Step S110: Acquire multimedia objects in the multimedia object list and extract relevant features of the multimedia objects;

[0047] Step S120: determining a target multimedia object based on relevant features of the multimedia object;

[0048] Step S130: extracting segment-related features of the target multimedia object;

[0049] Step S140: determining a target segment based on segment-related features;

[0050] Step S150: Prompt information in the target segment.

[0051] It can be seen from the steps of the above-mentioned information prompting method that the information prompting method provided by the exemplary embodiment of the present invention obtains the multimedia objects in the multimedia object list and extracts the relevant features of the multimedia objects. According to the relevant features of the multimedia objects, the target multimedia object for information prompting can be determined, and then the segment-related features are further extracted from the target multimedia object, and the target segment is determined according to the segment-related features, so as to perform information prompting in the target segment, so that information prompting can be performed at a more appropriate time, thereby improving the user experience and increasing the conversion effect of the information prompt.

[0052] The following describes in detail the steps of the information prompt method provided by the exemplary embodiment of the present invention in conjunction with specific embodiments:

[0053] In step S110 , multimedia objects in the multimedia object list are obtained, and features related to the multimedia objects are extracted.

[0054] In the exemplary embodiment of the present invention, the multimedia object may be an audio object, a video object, a text object, etc. In the following description of the embodiment, songs in the audio object are mainly used as an example for illustrative description, but this does not limit the multimedia objects in the exemplary embodiment of the present invention.

[0055] In actual applications, the above-mentioned multimedia object list can be a list corresponding to the playback entrance selected by the user. Taking songs as an example, the playback entrance can be a ranking list, a recently played list, or a playlist, etc. Among them, the playlist can be a selected playlist recommended to users daily by song playback platforms such as the Cloud Music platform, for example, a playlist provided according to different eras, languages, scenes, styles, emotions, themes, etc.

[0056] When the user enters the multimedia object list through the playback entrance, the information prompt method provided by the exemplary embodiment of the present invention can obtain the multimedia objects in the multimedia object list. For example, if the user enters from the most recent playback, the multimedia objects in the multimedia object list corresponding to the most recent playback can be obtained.

[0057] In an exemplary embodiment of the present invention, after the multimedia object is determined, features related to the multimedia object may be extracted.

[0058] In actual applications, different multimedia objects have different corresponding multimedia object-related features. Taking songs as an example, the corresponding multimedia object-related features may include: playback entry, average number of songs listened to, song tags, user song listening tags, user registration time, and user platform level. Among them, the average number of songs listened to can be the average number of songs listened to by the user in the recent period, for example, the average number of songs listened to by the user in the past week; song tags can be the language of the song, the style of the song, the emotional type of the song, etc.; user song listening tags refer to the tags of songs that the user often listens to; user registration time can be the time when the user registers a music playback application account, for example, the time when the user registers a cloud music platform app (Application) account; user platform level refers to the level of permissions enjoyed by the user in the music platform, which is usually related to the user's listening time.

[0059] In step S120, a target multimedia object is determined according to the multimedia object related features.

[0060] In an exemplary embodiment of the present invention, after extracting multimedia object-related features, the target multimedia object can be determined based on the multimedia object-related features. Specifically, in the process of determining the target multimedia object, the individual feature score corresponding to each multimedia object in the first preset number of multimedia objects in the multimedia object list can be determined based on the calculation logic of the associated features of each multimedia object; then, the total score of each multimedia object can be determined based on the individual feature score and the corresponding weight coefficient of each multimedia object-related feature; and the multimedia object with the highest total score is determined as the target multimedia object.

[0061] Next, we will continue to use songs as an example to illustrate the process of determining the target multimedia object.

[0062] Assume that after a user enters a multimedia object list from the playback portal, the multimedia object list contains 20 songs. Here, the target multimedia object is mainly determined from the first 10 songs. Table 1 shows how to determine the individual feature scores corresponding to each multimedia object based on the calculation logic.

[0063] Table 1

[0064]

[0065] The following explains Table 1. n1-n10 refers to the song's sequence number within the playlist, representing the first 10 songs in the playlist. The scores below n1-n10 are the individual feature scores for each song. These scores are determined primarily based on the calculation logic for each multimedia object's associated features.

[0066] For example, for the play entry feature, the calculation logic is to determine the score of each song based on the average number of songs a user listens to after entering the playlist through that play entry. If the average number of songs a user listens to after entering the playlist through the most recently played play entry is 6, then each of the first 6 songs will receive 1 point. Therefore, the play entry feature scores for the first 6 songs (n1-n6) in Table 1 are 1 point, and the scores for the remaining songs are 0.

[0067] For the average number of songs listened to, we can determine the score for each song based on the average number of songs the user listens to per session. For example, if the user listens to 12 songs per session, the first 10 songs can each receive 1 point, as shown in Table 1. However, if the user listens to 7 songs per session, we can assign 1 point to each of the first 7 songs, leaving the last 3 songs scoreless. This will be determined based on actual circumstances and will not be detailed here.

[0068] For the song tag feature, it is necessary to combine it with the user's song tag feature to determine the score of each song. Specifically, the score of each song can be determined based on the matching similarity score (0-1 points) between the user's song tag and the song tag. For details, please refer to the existing song similarity algorithm, which will not be described in detail here. The score of each song determined by the song similarity algorithm in the exemplary embodiment of the present invention is shown in Table 1.

[0069] For the user registration time feature, the score for each song can be determined based on the number of days since the user registered. For example, if the user registered within a week, the first two-thirds of the songs are scored zero; if the user registered more than a month ago, the first half of the songs are scored zero; and if the user registered more than three months ago, each song receives a score of 1. In Table 1, assuming the user registered within a week, the first two-thirds of the 10 songs are scored zero, meaning the first seven songs are scored zero, and the last three songs each receive a score of 1.

[0070] For the user platform level feature, the score for each song can be determined based on the user platform level and the total number of songs in the playlist. For example, if there are 10 user levels on the Cloud Music platform and the user level is x, the score can be determined using the formula x / 10*total number of songs in the playlist = k, where songs in the playlist with a score <= k each receive 1 point. In Table 1, the playlist has 20 songs and the user level is 5, so n1 to n10 each receive 1 point.

[0071] In an exemplary embodiment of the present invention, after determining individual feature scores, the total score for each multimedia object can be determined by combining the corresponding weight coefficients of each multimedia object-related feature. Assuming that the feature scores of each multimedia object-related feature are X1, X2, ..., Xn, and the weight coefficients of each feature are K1, K2, ..., Kn, then the total score for each multimedia object is X1*K1+X2*K2...+Xn*Kn. In Table 1, assuming that the weight coefficient of each multimedia object-related feature is 1, the total score obtained is as shown in Table 1.

[0072] It should be noted that as the number of subsequent samples increases, the weight coefficient of each multimedia object-related feature can be adjusted according to the user's historical conversion data. For example, according to the user's historical conversion data, if the serial numbers of the target multimedia objects that have been successfully converted to the user are all less than the user's average number of songs listened to, it means that in the user's daily listening process, the conversion rate of information prompts within the first m (m < n) songs is relatively high, indicating that the "average number of songs listened to" feature has a relatively large influence factor, and its weight coefficient needs to be increased. If the serial numbers of the target multimedia objects that have been successfully converted to the user are all greater than the number of songs listened to at each playback entrance, it means that the "playback entrance" feature has a relatively small influence factor, and its weight coefficient needs to be reduced.

[0073] In an exemplary embodiment of the present invention, the calculation logic can also be adjusted, and the calculation logic can be optimized according to the conversion results. For example, multiple users can be divided into two groups, a and b, for example, 50% of the users are randomly assigned to each group to ensure that the users are evenly distributed and not affected by other factors. For example, the calculation logic is adjusted based on the user registration time feature. Group a maintains the old logic unchanged, and group b does not score 1 / 2 of the songs for more than two weeks, and each song gets 1 point for more than 1 month. After observing for a period of time, the daily conversion data of groups a and b are compared, such as the number of purchased memberships. If the purchase data of group a is better than that of group b, it means that the adjustment logic of group b has a negative impact and can be discarded. On the contrary, if the data of group b is better than that of group a, it means that the adjusted calculation logic is better, and the calculation logic of group b can be used for the calculation logic of user registration time features.

[0074] In an exemplary embodiment of the present invention, after determining the total score of each multimedia object based on the individual feature scores and the corresponding weight coefficients of each multimedia object's related features, the multimedia object with the highest total score can be identified as the target multimedia object. For example, song n8 in Table 1 has a total score of 4.0, the highest score among the top 10 songs. Therefore, song n8 can be identified as the target song, i.e., the target multimedia object.

[0075] In practical applications, the pre-set number can be determined according to actual conditions, and the exemplary embodiment of the present invention does not impose any special limitation on this.

[0076] In step S130 , segment-related features of the target multimedia object are extracted.

[0077] In an exemplary embodiment of the present invention, after determining a target multimedia object, segment-related features can be extracted from the target multimedia object. A segment is a portion of a multimedia object with a certain characteristic. For example, for a song, its segments may include: prelude, verse, chorus, lead song, interlude, bridge, outro, etc. Among them, the prelude refers to the part at the beginning of the song without lyrics; the verse is the counterpart of the chorus and refers to the song's exposition; the chorus, originally "repetition song", refers to the climax of the song; the pre-chorus is used to connect the verse and chorus, and is generally short in length and has a different melody from the verse; the interlude is used to connect the previous chorus and the next verse, and generally has no lyrics; the bridge is used to connect the second chorus and the third chorus, and generally has lyrics. Bridge also means transition; and the outro is the part at the end of the song without lyrics. Generally, the prelude and outro each have only one section, while other parts may appear multiple times in a song.

[0078] In practical applications, segment-related features may be user behavior features in each segment of a multimedia object. In an exemplary embodiment of the present invention, segment-related features may include: playback duration in each dimension and number of segment type switching times in each dimension.

[0079] In actual applications, the playback time of each dimension can include on-site playback time, multimedia object playback time, and user playback time. Taking songs as an example, on-site playback time can be the playback time on the Cloud Music platform, multimedia object playback time can be the time the song is played, and user playback time can be the time the user listens to the song. The Cloud Music platform playback time refers to the average playback time of all Cloud Music platform users; song playback time refers to the average time a user listens to the song; and user listening time refers to the average time a user listens to a song in a single session.

[0080] In actual applications, the number of segment type switches in each dimension may include: the number of segment type switches in the site dimension, the number of segment type switches in the style dimension, and the number of segment type switches in the multimedia object dimension. Still taking songs as an example, the number of segment type switches in the site dimension may be the number of song cuts for each segment type under the cloud music platform dimension, the number of segment type switches in the style dimension may be the number of song cuts for each segment type under the genre dimension, and the number of segment type switches in the multimedia object dimension may be the number of song cuts for each segment type under the song dimension. Among them, the number of song cuts for each segment type under the cloud music platform dimension refers to the average number of song cuts for each segment type in the songs listened to by all users counted on the cloud music platform; the number of song cuts for each segment type under the genre dimension refers to, for example, the song cut situation in each segment when a user listens to a song of a certain genre on the cloud music platform; the number of song cuts for each segment type under the song dimension here refers to, for example, the average number of song cuts in each segment when a user listens to a target multimedia object on the cloud application platform.

[0081] It should be noted that the above-mentioned cloud music platform is only an exemplary description. In actual application, it can be any music playback platform, and the exemplary embodiments of the present invention are not limited to this.

[0082] Step S140: Determine the target segment according to the segment-related features.

[0083] In an exemplary embodiment of the present invention, determining the target segment based on segment-related characteristics may include: determining the user's maximum playback time based on the playback time of each of the above dimensions; and then determining the target segment within the user's maximum playback time range in the target multimedia object based on the number of segment type switching times of each of the above dimensions.

[0084] Specifically, determining the maximum playback time of a user based on the playback time of each dimension may include: determining the maximum playback time of the user based on the user playback time, the playback time within the site, and the playback time of the multimedia object, in combination with corresponding weights. The following example illustrates determining the maximum playback time of a user:

[0085] Assume that the playback durations for each dimension are: n for on-site playback, m for multimedia object playback, and l for user playback. The weights corresponding to each duration can be obtained through controlled experiments.

[0086] Specifically, in the process of obtaining weights through control experiments, first, the user experimental data can be randomly divided into different groups. Each group has a preset weight value, and at least one weight value is different between different groups. For example, as shown in Table 2, in groups t1 and t2, the weights corresponding to the multimedia object playback time are different; in groups t2 and t3, the weights corresponding to the site playback time are different; in groups t3 and t4, the weights corresponding to the user playback time are different.

[0087] Table 2

[0088] t1 t2 t3 t4 Playing time on the site 40％ 40％ 30％ 30％ Multimedia object playback duration 30％ 25％ 25％ 25％ User playback time 30％ 30％ 30％ 40％

[0089] If the user experimental data is evenly distributed between groups t1 and t2, the maximum user playback duration can be determined based on the weights in groups t1 and t2. Next, ensuring that other parameters for groups t1 and t2 are identical, target segments can be determined based on their respective maximum user playback durations. Users can be prompted with information during the target segments. Based on the final conversion results in groups t1 and t2, if the prompts are payment information, the conversion result can be the final proportion of paying users. If the number of paying users in group t1 is greater than that in group t2, this indicates that the weight of multimedia object playback duration has a greater impact. This weight can be increased and tuned until the number of paying users in group t1 no longer increases. This allows the weight of multimedia object playback duration to be determined as X2.

[0090] According to the above method, the user experimental data can be divided into t2 and t3 groups to determine the weight X1 of the in-site play time; the user experimental data can also be divided into t3 and t4 groups to determine the weight X3 of the user play time. No further details will be given here.

[0091] After determining the weights corresponding to the in-site playback time n, multimedia object playback time m, and user playback time l, the maximum user playback time a=n*X1+m*X2+l*X3 can be determined.

[0092] After determining the user's maximum playback duration a, we can identify the target segments within the target multimedia object within that maximum playback duration. Specifically, we can determine the segment type with the highest probability of being cut out within the target multimedia object's maximum playback duration based on the number of segment type switches within the multimedia object dimension, the number of segment type switches within the site dimension, and the number of segment type switches within the genre dimension, combined with their corresponding weights. The segment corresponding to the segment type with the highest probability of being cut out is then identified as the target segment.

[0093] The following uses songs as an example to explain the process of determining the target segment:

[0094] Assume that the number of switching times of the segment types in each dimension, that is, the number of song switching times of each segment type in the cloud music platform dimension, the number of song switching times of each segment type in the genre dimension, and the number of song switching times of each segment type in the song dimension are shown in Table 3.

[0095] Table 3

[0096] Prelude Main Song refrain Song Introduction Interlude Bridge Outro Cloud Music Platform Dimension n1 n2 n3 n4 n5 n6 n7 Genre dimension k1 k2 k3 k4 k5 k6 k7 Song Dimension m1 m2 m3 m4 m5 m6 m7

[0097] Then, the weights under the cloud music platform dimension, the weights under the music style dimension, and the weights under the song dimension are determined respectively. The weight determination process can also be obtained through a control experiment. Specifically, in the process of obtaining weights through a control experiment, first, the user experimental data can be randomly divided into different groups, each group has a preset weight value, and at least one weight value is different between different groups. For example, as shown in Table 4, in groups t5 and t6, the weights corresponding to the music style dimension are different; in groups t6 and t7, the weights corresponding to the cloud music platform dimension are different; in groups t7 and t8, the weights corresponding to the song dimension are different.

[0098] Table 4

[0099]

[0100]

[0101] Next, the average number of song cuts for each segment in different dimensions can be determined according to the preset weights in Table 4, and the type of segment with the maximum probability of being cut away within the user's maximum playback time in the target multimedia object can be determined. That is to say, it is necessary to first filter out the segments after the user's maximum playback time a, sort the average number of song cuts for the remaining segments, and obtain the segment type with the largest average number of song cuts as the segment type with the maximum probability of being cut away, and determine the segment corresponding to the segment type with the maximum probability of being cut away as the target segment for information prompting.

[0102] In practical applications, the weights in Table 4 can be optimized and adjusted based on actual experimental results, which will not be described in detail here. After the final weights are determined through optimization, the final maximum probability cut-off segment type can be determined.

[0103] In actual applications, there may be multiple segments corresponding to the maximum probability of cutting out the segment type determined. For example, the maximum probability of cutting out the segment type determined in the previous step is the chorus, and for the target multimedia object, there are two corresponding chorus segments. In this case, it is necessary to further determine the final target segment from these two chorus segments.

[0104] In an exemplary embodiment of the present invention, when there are multiple segments corresponding to the maximum probability cut-away segment type; the process of determining the segment corresponding to the maximum probability cut-away segment type as the target segment may include: first determining the similarity between the label corresponding to each segment corresponding to the maximum probability cut-away segment type and the user playback label; and determining the segment with the greatest similarity as the target segment.

[0105] In the process of determining the similarity, a variety of different similarity algorithms can be used for determination. This exemplary embodiment takes the cosine similarity algorithm as an example to illustrate the above similarity determination process as follows:

[0106] Assume that the segment type with the highest probability of being cut off is the chorus, and for the target multimedia object, there are two corresponding chorus segments: chorus segment 1 and chorus segment 2.

[0107] First, determine the tag segmentation:

[0108] listA: Chorus 1 tag

[0109] listB: Chorus 2 Tags

[0110] listC: user's song listening tags

[0111] If list A = ['pure sound', 'light music', 'soothing', 'popular', 'healing',

[0112] 'Quiet']

[0113] list B = ['Exciting', 'Healing', 'Chinese', 'Popular', 'Travel', 'Touching']

[0114] list C = ['Mandarin', 'Pop', 'Pure Sound', 'Quiet', 'Folk', 'Classical']

[0115] Next, list all the words:

[0116] Put listA and listC in a set1, and we get:

[0117] set1 = {'pure sound', 'light music', 'soothing', 'popular', 'healing',

[0118] 'Quiet', 'Mandarin', 'Folk', 'Classical'}

[0119] Put listB and listC in a set2, and we get:

[0120] Set2 = {'Exciting', 'Healing', 'Chinese', 'Pop', 'Travel', 'Touching', 'Pure Sound', 'Quiet', 'Folk', 'Classical'}

[0121] Convert the above set1 and set2 into dict respectively, where the key is the word in the set and the value is the position where the word appears in the set, that is, in the form of 'pure sound':1.

[0122] dict1 = {'pure sound': 0, 'light music': 1, 'soothing': 2, 'popular': 3,

[0123] 'Healing': 4, 'Quiet': 5, 'Mandarin': 6, 'Folk': 7, 'Classical': 8}

[0124] dict2 = {'Exciting': 0, 'Healing': 1, 'Chinese': 2, 'Popular': 3, 'Travel': 4, 'Emotional': 5, 'Pure Sound': 6, 'Quiet': 7, 'Folk': 8, 'Classical': 9}

[0125] Next, encode the list:

[0126] Convert each word to the position in the set, and convert dict1 and dict2 respectively to:

[0127] dict1:

[0128] list A = ['pure sound', 'light music', 'soothing', 'popular', 'healing',

[0129] 'Quiet']

[0130] list C = ['Mandarin', 'Pop', 'Pure Sound', 'Quiet', 'Folk', 'Classical']

[0131] After conversion

[0132] list A=[0,1,2,3,4,5]

[0133] list C=[2, 3, 0, 5, 7, 8]

[0134] dtic2:

[0135] list B = ['Exciting', 'Healing', 'Chinese', 'Popular', 'Travel', 'Touching']

[0136] list C = ['Mandarin', 'Pop', 'Pure Sound', 'Quiet', 'Folk', 'Classical']

[0137] After conversion

[0138] list B=[0,1,2,3,4,5]

[0139] list C=[2, 3, 6, 7, 8, 9]

[0140] Further, perform oneHot encoding on listcode:

[0141] That is, the number of times each word appears is calculated, and the results after oneHot numbering are as follows:

[0142] dict1 = {'pure sound': 0, 'light music': 1, 'soothing': 2, 'popular': 3,

[0143] The number of times the participles in list A and list C appear in the following sentences: 'cure': 4, 'quiet': 5, 'Chinese': 6, 'folk': 7, 'classical': 8

[0144] listAcodeOneHot=[1,1,1,1,1,1,0,0,0]

[0145] listCcodeOneHot=[1,0,0,1,0,1,1,1,1]

[0146] dict2 = {'exciting': 0, 'healing': 1, 'Chinese': 2, 'popular': 3, 'travel': 4, 'touching': 5, 'pure sound': 6, 'quiet': 7, 'folk': 8, 'classical': 9} The number of times the participle appears in listB and listC respectively:

[0147] listBcodeOneHot=[1,1,1,1,1,1,0,0,0,0]

[0148] listCcodeOneHot=[0,0,1,1,0,0,1,1,1,1]

[0149] Finally, the cosine similarity is calculated as follows:

[0150] After obtaining the word frequency vectors of the two tags, it becomes a matter of calculating the cosine value of the angle between the two vectors. The larger the value, the higher the similarity.

[0151] The calculation formula is:

[0152]

[0153] List A, C similarity value:

[0154]

[0155] Similarity values ​​of list B and C:

[0156]

[0157] According to the cosine similarity, the similarity between list A and C is higher than that between list B and C, indicating that the chorus segment 1 corresponding to list A is the best position, and the chorus segment 1 can be determined as the target segment.

[0158] Step S150: Prompt information in the target segment.

[0159] In an exemplary embodiment of the present invention, after a target segment is determined, a prompt may be provided to the user upon playback of the target segment, prompting the user to perform operations related to obtaining permission to play subsequent segments. If the prompt information is payment information, the user may obtain membership privileges or continue listening after paying. The prompt information is not specifically limited herein.

[0160] The information prompt method provided by the exemplary embodiment of the present invention determines the target multimedia object through the relevant features of the multimedia object, and then determines the target segment based on the relevant features of the segment related to the user behavior, so as to provide information prompts when playing the target segment. This is equivalent to providing information prompts based on the characteristics of each user, so that the timing of information prompts can be better grasped, thereby improving the user experience and increasing the user conversion rate.

[0161] Exemplary devices

[0162] After introducing the information prompt method of the exemplary embodiment of the present invention, Figure 2 The information prompt device according to an exemplary embodiment of the present invention is described, wherein the device embodiment part can inherit the relevant description in the method embodiment, so that the device embodiment can obtain the support of the relevant specific description in the method embodiment.

[0163] refer to Figure 2 According to an exemplary embodiment of the present invention, the information prompting device 200 may include: a first feature extraction module 210, a target multimedia object determination module 220, a second feature extraction module 230, a target segment determination module 240 and an information prompting module 250.

[0164] Specifically, the first feature extraction module 210 can be used to extract multimedia object-related features when obtaining multimedia objects in the multimedia object list; the target multimedia object determination module is used to determine the target multimedia object based on the multimedia object-related features; the second feature extraction module is used to extract segment-related features of the target multimedia object; the target segment determination module is used to determine the target segment based on the segment-related features; and the information prompt module is used to provide information prompts in the target segment.

[0165] In some embodiments of the present invention, the target multimedia object determination module 220 can be used to determine the individual feature score corresponding to each multimedia object in the first preset number of multimedia objects in the multimedia object list based on the calculation logic of the relevant features of each multimedia object; then determine the total score of each multimedia object based on the individual feature score and the corresponding weight coefficient of the relevant features of each multimedia object; and determine the multimedia object with the highest total score as the target multimedia object.

[0166] In some embodiments of the present invention, the segment-related features include: the playback duration of each dimension and the number of segment type switching times of each dimension.

[0167] In some embodiments of the present invention, the target segment determination module 240 can be used to determine the maximum playback time of the user based on the playback time of each dimension; and determine the target segment within the maximum playback time range of the user in the target multimedia object based on the number of segment type switches in each dimension.

[0168] In some embodiments of the present invention, the target segment determination module 240 may be configured to determine the maximum playback duration of the user based on the user playback duration, the on-site playback duration, and the multimedia object playback duration, in combination with corresponding weights.

[0169] In some embodiments of the present invention, the target segment determination module 240 can be used to determine the segment type with the highest probability of being cut away within the maximum playback time of the user in the target multimedia object based on the number of segment type switches in the multimedia object dimension, the number of segment type switches in the site dimension, and the number of segment type switches in the style dimension, combined with the corresponding weights; and determine the segment corresponding to the segment type with the highest probability of being cut away as the target segment.

[0170] In some embodiments of the present invention, when there are multiple segments corresponding to the segment type with the maximum probability of being cut away, the target segment determination module 240 can be used to determine the similarity between the label corresponding to each segment corresponding to the segment type with the maximum probability of being cut away and the user playback label; and the segment with the greatest similarity is determined as the target segment.

[0171] In some embodiments of the present invention, the target multimedia object determination module 220 may be configured to adjust the weight coefficient of each multimedia object-related feature according to the user's historical conversion data.

[0172] In some embodiments of the present invention, multimedia object related features include: playback entry, average number of songs listened to, song tags, user song tags, user registration time, and user platform level or more.

[0173] In addition, other specific details of the embodiments of the present invention have been described in detail in the embodiment of the invention of the above method and will not be repeated here.

[0174] The information prompt device provided by the exemplary embodiment of the present invention determines the target multimedia object through the relevant features of the multimedia object, and then determines the target segment based on the relevant features of the segment related to the user behavior, so as to provide information prompts when playing the target segment. This is equivalent to providing information prompts based on the characteristics of each user, so that the timing of information prompts can be better grasped, thereby improving the user experience and increasing the user conversion rate.

[0175] Exemplary Storage Media

[0176] refer to Figure 3 A storage medium according to an exemplary embodiment of the present invention is described.

[0177] like Figure 3 As shown, a program product 300 for implementing the above method according to an embodiment of the present invention is described. The program product 300 may be a portable compact disk read-only memory (CD-ROM) and include program code, and may be run on a device such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0178] The program product can be implemented in any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0179] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0180] The program code contained on the readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RE, etc., or any suitable combination of the foregoing.

[0181] The program code for performing the operations of the present invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, and the like, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (FAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0182] Exemplary electronic devices

[0183] refer to Figure 4 An electronic device according to an exemplary embodiment of the present invention is described.

[0184] Figure 4 The electronic device 400 shown is merely an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0185] like Figure 4 As shown, electronic device 400 is implemented as a general-purpose computing device. Components of electronic device 400 may include, but are not limited to, at least one processing unit 410, at least one storage unit 420, a bus 430 connecting various system components (including storage unit 420 and processing unit 410), and a display unit 440.

[0186] The storage unit stores program code that can be executed by processing unit 410, causing processing unit 410 to perform the steps according to various exemplary embodiments of the present invention described in the "Exemplary Methods" section above. For example, processing unit 410 may perform the method steps shown in the figure.

[0187] The storage unit 420 may include a volatile storage unit, such as a random access memory unit (RAM) 4201 and / or a cache memory unit 4202 , and may further include a read-only memory unit (ROM) 4203 .

[0188] The storage unit 420 may also include a program / utility 4204 having a set (at least one) of program modules 4205, such program modules 4205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0189] The bus 430 may include a data bus, an address bus, and a control bus.

[0190] The electronic device 400 can also communicate with one or more external devices 470 (e.g., a keyboard, a pointing device, a Bluetooth device, etc.), and such communication can be performed via an input / output (I / O) interface 450. The electronic device 400 also includes a display unit 440, which is connected to the input / output (I / O) interface 450 for display. In addition, the electronic device 400 can also communicate with one or more networks (e.g., a local area network (FAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 460. As shown, the network adapter 460 communicates with other modules of the electronic device 400 via a bus 430. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0191] It should be noted that although several modules or submodules of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more units / modules described above may be embodied in a single unit / module. Conversely, the features and functions of a single unit / module described above may be further divided and embodied by multiple units / modules.

[0192] Furthermore, although the operations of the method of the present invention are described in a particular order in the accompanying drawings, this does not require or imply that these operations must be performed in this particular order, or that all illustrated operations must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0193] While the spirit and principles of the present invention have been described with reference to several specific embodiments, it should be understood that the present invention is not limited to the specific embodiments disclosed, and the division into various aspects does not mean that the features of these aspects cannot be combined to benefit. Such division is merely for convenience of expression. The present invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. An information prompting method, characterized in that: include: Obtaining multimedia objects in the multimedia object list and extracting relevant features of the multimedia objects; determining a target multimedia object according to the relevant features of the multimedia object; Extracting segment-related features of the target multimedia object, where the segment is a portion of the target multimedia object with specific characteristics, and the segment-related features include: playback duration of each dimension and number of segment type switching times of each dimension; determining a target segment according to the segment-related features; An information prompt is provided in the target segment to prompt the user to perform relevant operations to obtain the permission to play subsequent segments.

2. The method according to claim 1, characterized in that The determining the target multimedia object according to the multimedia object related features includes: determining, according to calculation logic of the relevant features of each multimedia object, a single feature score corresponding to each multimedia object in a first preset number of multimedia objects in the multimedia object list; Then, determining a total score of each multimedia object according to the single feature score and the corresponding weight coefficient of each multimedia object-related feature; The multimedia object with the highest total score is determined as the target multimedia object.

3. The method according to claim 1, characterized in that The determining of the target segment according to the segment-related features includes: Determine the maximum playback time of the user based on the playback time of each dimension; The target segment within the maximum playback duration of the user in the target multimedia object is determined according to the number of segment type switching times in each dimension.

4. The method according to claim 3, characterized in that Determining the maximum playback time of the user according to the playback time of each dimension includes: The maximum playback time of the user is determined based on the user playback time, the site playback time and the multimedia object playback time, combined with the corresponding weights.

5. The method according to claim 3, characterized in that The determining, based on the number of segment type switching times in each dimension, the target segment within the maximum playback duration of the user in the target multimedia object comprises: Determine the maximum probability of switching the segment type within the target multimedia object within the maximum playback duration of the user based on the number of segment type switches in the multimedia object dimension, the number of segment type switches in the site dimension, and the number of segment type switches in the genre dimension, combined with their corresponding weights; The segment corresponding to the segment type with the maximum probability is cut off and determined as the target segment.

6. The method according to claim 5, characterized in that In a case where there are multiple segments corresponding to the segment type with the maximum probability of being cut away, determining the segment corresponding to the segment type with the maximum probability of being cut away as the target segment includes: Determine the similarity between the label corresponding to each segment corresponding to the segment type with the maximum probability of being cut off and the user's playback label; The segment with the greatest similarity is determined as the target segment.

7. The method according to claim 2, characterized in that The method further comprises: The weight coefficient of each multimedia object-related feature is adjusted according to the user's historical conversion data.

8. The method according to claim 1, characterized in that The multimedia object related features include: one or more of: playback entry, average number of songs listened to, song tags, user song listening tags, user registration time, and user platform level.

9. An information prompting device, characterized in that: include: A first feature extraction module is used to extract features related to multimedia objects when obtaining multimedia objects in the multimedia object list; a target multimedia object determining module, configured to determine a target multimedia object based on the multimedia object related features; a second feature extraction module, configured to extract segment-related features of the target multimedia object, wherein the segment is a portion of the target multimedia object having specific characteristics, and the segment-related features include: playback duration of each dimension and number of segment type switching times of each dimension; a target segment determination module, configured to determine a target segment based on the segment-related features; The information prompt module is used to provide information prompts for the target segment to prompt the user to perform relevant operations to obtain the subsequent segment playback permission.

10. The device according to claim 9, characterized in that The target multimedia object determination module is used to: determining, according to calculation logic of the relevant features of each multimedia object, a single feature score corresponding to each multimedia object in a first preset number of multimedia objects in the multimedia object list; Then, determining a total score of each multimedia object according to the single feature score and the corresponding weight coefficient of each multimedia object-related feature; The multimedia object with the highest total score is determined as the target multimedia object.

11. The device according to claim 9, characterized in that The target segment determination module is used for: Determine the maximum playback time of the user based on the playback time of each dimension; The target segment within the maximum playback duration of the user in the target multimedia object is determined according to the number of segment type switching times in each dimension.

12. The device according to claim 11, characterized in that The target segment determination module is used to determine the maximum playback time of the user according to the user playback time, the site playback time and the multimedia object playback time, combined with the corresponding weights.

13. The device according to claim 11, characterized in that The target segment determination module is used for: Determine the maximum probability of switching the segment type within the target multimedia object within the maximum playback duration of the user based on the number of segment type switches in the multimedia object dimension, the number of segment type switches in the site dimension, and the number of segment type switches in the genre dimension, combined with their corresponding weights; The segment corresponding to the segment type with the maximum probability is cut off and determined as the target segment.

14. The device according to claim 13, characterized in that In the case where there are multiple segments corresponding to the maximum probability cut-away segment type; the target segment determination module is used to determine the similarity between the label corresponding to each segment corresponding to the maximum probability cut-away segment type and the user playback label; and determine the segment with the greatest similarity as the target segment.

15. The device according to claim 10, characterized in that The target multimedia object determination module is configured to adjust the weight coefficient of each multimedia object-related feature according to the user's historical conversion data.

16. The device according to claim 9, characterized in that The multimedia object related features include: one or more of: playback entry, average number of songs listened to, song tags, user song listening tags, user registration time, and user platform level.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

18. An electronic device, characterized in that: include: processor; as well as a memory for storing executable instructions of the processor; The processor is configured to perform the method according to any one of claims 1 to 8 by executing the executable instructions.

Images