Customer broadcasting information customization method and device, intelligent equipment and storage medium

Abstract

The invention provides a broadcast information customization method and device, intelligent equipment and a storage medium, and belongs to the technical field of broadcast information customization, and the method comprises the steps: generating a random broadcast type based on a plurality of preference broadcast types determined by a user, the preference broadcast types being preference types set when the user uses broadcast for the first time; a first weight is allocated to the plurality of preference broadcaster types to obtain first preference type data, a second weight is allocated to the random broadcaster type to obtain first random type data, and the first weight is greater than the second weight; and customizing broadcasting information for the user based on the first preference type data and the first random type data. According to the broadcast information customization method and device, the intelligent equipment and the storage medium provided by the invention, the user experience can be improved.

Description

Technical Field

[0001] The present disclosure belongs to the technical field of podcast information customization, and more specifically, to a podcast information customization method, apparatus, smart device and storage medium. Background Art

[0002] With the rapid development of Internet technology and the advent of the information age, podcasts, as a new form of digital audio media, have gradually become an important channel for people to obtain information and entertainment. However, faced with a huge amount of podcast resources, how to accurately recommend podcast information that meets the personalized needs of users has become a major challenge facing current podcast platforms.

[0003] Traditional podcast recommendation methods often rely on users' active search and subscription behaviors. Although this method can meet users' specific needs to a certain extent, it lacks initiative and diversity, resulting in a poor user experience. Summary of the invention

[0004] The purpose of the present disclosure is to provide a podcast information customization method, apparatus, smart device and storage medium to improve user experience.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for customizing podcast information is provided, wherein the method comprises: generating a random podcast type based on a plurality of preferred podcast types determined by the user, the preferred podcast type being a preferred type set by the user when using the podcast for the first time; Assigning a first weight to a plurality of preferred podcast types to obtain first preferred type data, assigning a second weight to a random podcast type to obtain first random type data, the first weight being greater than the second weight; Podcast information is customized for the user based on the first preference type data and the first random type data.

[0006] A second aspect of an embodiment of the present disclosure provides a podcast information customization device, including: An information customization setting module, used to generate a random podcast type based on multiple preferred podcast types determined by the user, where the preferred podcast type is the preferred type set by the user when using the podcast for the first time; A weight assignment module, configured to assign a first weight to a plurality of preferred podcast types to obtain first preferred type data, and assign a second weight to a random podcast type to obtain first random type data, wherein the first weight is greater than the first weight; The information customization module is used to customize podcast information for users based on the first preference type data and the first random type data.

[0007] According to a third aspect of an embodiment of the present disclosure, a smart device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the above-mentioned podcast information customization method when executing the computer program.

[0008] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned podcast information customization method are implemented.

[0009] The beneficial effects of the podcast information customization method, apparatus, smart device, and storage medium provided by the embodiments of the present disclosure are as follows: the embodiments of the present disclosure generate random podcast types based on the preferred podcast types initially set by the user, which not only respects the user's existing interests, but also introduces potentially related content, thereby broadening the user's audio-visual range. At the same time, by assigning different weights to the preferred podcast types and random podcast types, it ensures that the recommended content can not only closely fit the user's preferences, but also introduce fresh elements in moderation, thereby avoiding the singleness and limitations of the recommended content. This makes the customization of podcast information more personalized, comprehensive, and exploratory, greatly improving the user's listening experience and willingness to continue using the podcast platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A flowchart of a podcast information customization method provided by an embodiment of the present disclosure; Figure 2 A structural block diagram of a podcast information customization device provided by an embodiment of the present disclosure; Figure 3 A schematic block diagram of a smart device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0012] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0013] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0014] Please refer to Figure 1 , Figure 1 A flowchart of a podcast information customization method provided in an embodiment of the present disclosure, the method comprising: S101: Generate a random podcast type based on multiple preferred podcast types determined by a user, where the preferred podcast type is a preferred type set by the user when using a podcast for the first time.

[0015] In this embodiment, when users use the podcast service for the first time, they can set their preferred podcast types. These preferred types may include but are not limited to music, technology, news, education, entertainment, etc. These settings reflect the basic interests and needs of users.

[0016] On this basis, this embodiment can generate random podcast types according to the preferred podcast types determined by the user.

[0017] For example, if the user's preference is "Technology Review", the randomly generated podcast type can be "Technology Frontiers", which aims to expand the related fields that the user is potentially interested in but may not have directly selected at the beginning, thereby enriching the diversity of recommendations and avoiding the recommendation content being too limited to the few clear preferences set by the user at the beginning.

[0018] In this embodiment, a random podcast type may be generated based on a method of associating multiple preferred podcast type content tags.

[0019] First, define podcast content tags for your podcast content.

[0020] Podcast platforms can add rich and detailed content tags to each podcast information. These tags cover multiple dimensions such as themes (such as "history", "technology", "food", etc.), styles ("serious", "humorous", "professional depth", etc.), and forms ("interview", "single-person narration", "plot interpretation", etc.). For example, a podcast about "ancient Roman history" can be tagged with themes such as "history", "humanities", "ancient civilization", "European history", as well as style tags such as "serious explanation" and form tags such as "single-person narration".

[0021] Second, obtain a plurality of preference tags for preferred podcast types.

[0022] After the user sets the preferred podcast type, this embodiment can extract the tags corresponding to these preferred types. For example, if the user prefers "cultural and historical interpretation", the relevant tags may be "history", "culture", "interpretation" and so on.

[0023] Finally, random podcast genres are generated based on semantic similarity.

[0024] By calculating the similarity between the user's preference tags and the podcast content tags. For example, the "history" tag often appears with tags such as "archaeology", "cultural relics", and "ancient architecture", which means that these tags are closely related. This embodiment can filter out podcast types with these related tags and recommend them to users as random podcast types. For example, "archaeological discovery sharing" can become a random type generated based on the user's "cultural history interpretation" preference.

[0025] S102: Assigning a first weight to a plurality of preferred podcast types to obtain first preferred type data, and assigning a second weight to a random podcast type to obtain first random type data, wherein the first weight is greater than the second weight.

[0026] In this embodiment, a first weight is assigned to a plurality of preferred podcast types to obtain first preference type data. Since the plurality of preferred podcast types are types that the user actively indicates interest in, the first weight is greater than the second weight assigned to the random podcast type. A higher weight means that the user's preference will be given priority when customizing podcast information. For example, assuming that the preference type of "Cultural and Historical Interpretation" is assigned a weight of 0.6 (first weight), it indicates that it is critical during customization, and its corresponding podcast content should be considered in particular to satisfy the user's existing preferences.

[0027] Assign the second weight to the random podcast type to get the first random type data. The relatively small weight indicates that it is more of a supplementary element, used to moderately broaden the recommendation range and introduce some fresh content that users may be interested in but not actively choose at the beginning. For example, the randomly generated type "Science and Technology Updates" is assigned a weight of 0.2 (the second weight), which has less influence in the overall customization than the preference type actively set by the user.

[0028] S103: Customizing podcast information for the user based on the first preference type data and the first random type data.

[0029] In this embodiment, podcast information is customized for the user based on the obtained first preference type data and first random type data. This embodiment can comprehensively analyze these two types of data with different weights, and select content that meets the corresponding weight ratio from the podcast resource library for recommendation.

[0030] For example, in the podcast list that is finally recommended to users, most of the content is highly relevant to the podcast type that the user prefers, and it occupies a larger space according to its higher weight; at the same time, some content belonging to the random podcast type will also be interspersed, and it occupies a relatively smaller space according to its lower weight. In this way, it not only ensures that the user's existing clear interests are met, but also appropriately guides the user to discover some new areas that may be of interest, making the podcast recommendation more personalized, comprehensive and exploratory, and improving the user's experience of listening to podcasts.

[0031] From the above, it can be concluded that this embodiment generates random podcast types based on the preferred podcast types initially set by the user, which not only respects the user's existing interests, but also introduces potentially related content, broadening the user's audio-visual range. At the same time, by assigning different weights to the preferred podcast types and random podcast types, it ensures that the recommended content can not only closely fit the user's preferences, but also introduce fresh elements appropriately, avoiding the singleness and limitations of the recommended content. This makes the customization of podcast information more personalized, comprehensive and exploratory, greatly improving the user's listening experience and willingness to continue using the podcast platform.

[0032] In one embodiment of the present disclosure, it also includes: updating a first weight corresponding to the preferred podcast type based on the playback duration of the podcast information corresponding to the preferred podcast type to obtain second preferred type data; Based on the playback duration of the podcast information corresponding to the random podcast type, a second weight corresponding to the random podcast type is updated to obtain second random type data; The podcast information is customized for the user based on the second preference type data and the second random type data.

[0033] In this embodiment, the user's behavior of listening to podcast information is continuously tracked, and the duration of the podcast information corresponding to each preferred podcast type is paid attention to. For example, the preferred podcast types set by the user include "Cultural History Interpretation" and "Technology Review". This embodiment can record the data such as the duration of each playback and the cumulative playback duration when the user listens to these two types of podcast content.

[0034] Playing time can reflect the user's actual interest in the type of podcast content. If the podcast information corresponding to a preferred podcast type is played for a long time, it means that the user is more interested in this type of content and has invested more time in listening to it. In this case, this type can be given a higher weight in subsequent recommendations to make it more prominent. On the contrary, if the playing time is short, the weight may need to be appropriately reduced.

[0035] For example, the original weight of the "Cultural History Interpretation" podcast was 0.6, but after a period of monitoring, it was found that the user's playback time continued to increase. In this embodiment, its weight can be increased to 0.7, forming the second preference type data, which means that when customizing podcast information subsequently, more emphasis should be placed on recommending this type of content.

[0036] In this embodiment, for random podcast types, this embodiment can also collect the playback time of the corresponding podcast information. For example, the random podcast type "Technology Frontier Dynamics" previously generated based on the "Technology Review" preference can record the time the user listens to this type of podcast.

[0037] If the playback time of "Frontiers of Science and Technology" is relatively long, it means that users are more receptive to this type of randomly recommended content and are interested in further exploration, so its weight can be appropriately increased.

[0038] For example, by increasing the initial weight from 0.2 to 0.3, a second random type of data is formed, which makes it account for a larger proportion and appear more frequently in subsequent customized recommendations; if the playback time is short, it indicates that the user is not very interested in it, and the weight can be reduced accordingly, reducing the chance of it appearing in the recommendation list.

[0039] After obtaining the second preference type data and the second random type data, this embodiment once again comprehensively analyzes these data after adjusting the weights by the play time. At this time, the customized podcast information is no longer based solely on the initially set weights, but is screened and recommended based on the weights updated based on the user's actual listening behavior feedback.

[0040] For example, after the weight of "Cultural and Historical Interpretation" is increased, the proportion of related podcasts in the recommendation list will further increase. If the weight of "Frontiers of Science and Technology" changes, its frequency of appearance and recommendation priority will also change accordingly, making podcast recommendations more in line with users' current real interests and listening habits, and achieving more accurate and user-satisfying personalized recommendation effects.

[0041] From the above, it can be concluded that this embodiment can accurately capture the dynamic changes of user interests by updating the weights of the preferred podcast type and the random podcast type based on the playback time of the corresponding podcast information. The weight of the preferred podcast type with a longer playback time is increased, which can enhance the satisfaction of the user's core interests and ensure the high-quality recommendation of core favorite content; and the weight of the random podcast type is adjusted according to the playback time, which can more accurately screen out the content that the user is potentially interested in and optimize the recommendation ratio. In this way, the podcast recommendation can deeply fit the actual needs and preferences of the user, greatly improving the accuracy and personalization of the recommendation.

[0042] In one embodiment of the present disclosure, before customizing podcast information for the user based on the second preference type data and the second random type data, the method further includes: generating a preference data set based on the second preference type data, and generating a random data set based on the second random type data; The preference data set is updated based on the first weight corresponding to the second preference type data, and the random data set is updated based on the second weight corresponding to the second random type data.

[0043] In this embodiment, the second preference type data can be obtained by updating the first weight originally corresponding to the podcast information based on the playback time of the preferred podcast type. The preference data set can be an integration of various types of information related to the user's preferred podcast type, such as the name, content introduction, host information and many other elements of the podcast program that meets the preference type, to form a relatively concentrated data set, which is convenient for subsequent operation and management. Similarly, a random data set is generated based on the second random type data, that is, the corresponding information related to the random podcast type is collected and organized to form an independent data set.

[0044] In this embodiment, since the first weight reflects the importance of the preferred podcast type in the current situation, the update process can adjust the relevant parameters or sorting of each element in the preference data set according to the size of this weight.

[0045] For example, if the preference type of "Cultural and Historical Interpretation" has a higher weight after adjustment for playback time, then in the preference data set, the information related to the podcast program belonging to the "Cultural and Historical Interpretation" type will be given a higher priority ranking, a higher recommendation score, etc., making it more likely to be screened out and displayed to users in the subsequent customization of podcast information, thereby highlighting the importance of user preferences and the interest tendencies reflected by the user's actual listening performance.

[0046] Similarly, for random data sets, updates are implemented according to the second weight corresponding to the second random type data. Because the second weight reflects the importance of the random podcast type, the program information corresponding to the random podcast type with an increased weight will have a higher display position and more prominent display opportunities in the data set, so that the randomly recommended content is more in line with the user's current acceptance of this type of potential interest content, and its influence in the overall recommendation system is reasonably allocated through weight.

[0047] From the above, it can be concluded that this embodiment can effectively integrate and classify information related to different podcast types by generating preference data sets and random data sets, making the data more organized. By adjusting the data set based on the updated weights, it can accurately highlight the status of content with high user attention in the preferred podcast type in the data set based on the dynamic changes in interests reflected by the user's actual listening behavior, and reasonably optimize the presentation of content related to the random podcast type. Greatly improve the accuracy and personalization of podcast recommendations and the efficiency of data processing.

[0048] In one embodiment of the present disclosure, updating the preference data set based on the first weight corresponding to the second preference type data, and updating the random data set based on the second weight corresponding to the second random type data, includes: In response to a first weight corresponding to the second preference type data in the preference data set being less than a first preset weight, storing the corresponding preference type data in the preference data set into the random data set; In response to a second weight corresponding to the second random type data in the random data set being greater than or equal to the first preset weight, the corresponding random type data in the random data set is stored in the preference data set.

[0049] In this embodiment, a first preset weight can be set, and the first preset weight serves as a key limit for measuring the change in importance of the preferred podcast type. When it is detected that the first weight corresponding to the second preference type data in the preference data set is less than the first preset weight, it means that the importance of certain podcast types that were originally preferred by the user has dropped to a certain extent after the weights are dynamically adjusted in combination with factors such as the playback duration. In this embodiment, the corresponding preference type data in the preference data set can be stored in the random data set.

[0050] For example, suppose "music appreciation" was originally the user's preferred podcast type. As the user's listening behavior changes, its corresponding first weight is updated to be less than the first preset weight, indicating that the user's current interest in this type of content is weaker than other preferred types. At this time, the data related to "music appreciation" can be transferred from the preference data set to the random data set, so that it can be recommended as a random content in the subsequent recommendations. It is no longer recommended based on the user's active preference, but is treated as a potential choice to expand the user's interests.

[0051] For the random data set, this embodiment can refer to the first preset weight for judgment. When it is found that the second weight corresponding to the second random type data in the random data set is greater than or equal to the first preset weight, it indicates that the importance of these podcast types originally recommended as random recommendations has been increased to a higher level after the weights are updated according to the playback time. This embodiment can store these random type data corresponding to the random data set in the preference data set.

[0052] For example, if the second weight of the random podcast type "Technology Frontiers" generated based on the user's "Technology Review" preference is greater than or equal to the first preset weight, it means that the user has shown a high interest in this type of content, which is equally important as those preference types that the user actively sets. Therefore, the data related to "Technology Frontiers" is transferred from the random data set to the preference data set. When customizing podcast information later, it can be regarded as the user's core preferred content and recommended in a focused manner to better meet the user's current interest needs.

[0053] From the above, it can be concluded that this embodiment can keenly capture the dynamic evolution of user interests by transferring data between data sets based on the comparison of weights with preset weights. When the weight of the preference type decreases, it is transferred to the random data set to avoid over-recommendation of content in which the user's interest has declined, and optimize the allocation of recommendation resources; and when the weight of the random type increases, it is moved to the preference data set, which can timely promote the content that the user has newly developed a strong interest in to the core recommendation level. This allows the customized content of the podcast to accurately adapt to the trajectory of user interest changes, improve the fit between the recommended content and the actual needs of the user, and enhance the user experience.

[0054] In one embodiment of the present disclosure, it also includes: In response to the second weight corresponding to the random type data in the random data set being less than the second preset weight, deleting the corresponding random type data in the random data set; The second preset weight is smaller than the first preset weight.

[0055] In this embodiment, two different weight limits are set, namely, a first preset weight and a second preset weight, and the second preset weight is less than the first preset weight. The two different weights can be understood as reference standards for measuring the importance of random type data in the random data set. This embodiment can continue to pay attention to the second weight corresponding to each random type data in the random data set, and by comparing it with the second preset weight, determine whether the corresponding random type data needs to be deleted and no longer recommended.

[0056] When it is found that the second weight corresponding to a certain random type of data in the random data set is less than the second preset weight, it means that this type of randomly recommended podcast type has very low appeal to users, and users pay little attention to or rarely listen to the related podcast content.

[0057] For example, the second weight corresponding to the random podcast type "Niche Art Sharing" is always less than the second preset weight after multiple updates, indicating that it has not played a positive role in expanding user interests and enriching recommendation diversity. In order to streamline the random data set and avoid it being filled with too much content that is of no value to users, and to optimize the system's data storage and data processing efficiency during subsequent recommendations, this embodiment can delete the corresponding random type data such as "Niche Art Sharing" in the random data set, so that it no longer participates in the subsequent podcast information customization process, thereby allowing the entire podcast recommendation mechanism to focus on podcast types that are more in line with user interests and more valuable, thereby improving the accuracy and effectiveness of the overall recommendation.

[0058] From the above, it can be concluded that this embodiment makes the data set more refined by deleting low-weight random type data to avoid worthless information from flooding the random data set. This not only improves data processing efficiency, but also allows recommendations to focus on podcast types that users are truly interested in or have potential interests in, thereby enhancing recommendation accuracy and personalization and improving user experience.

[0059] In one embodiment of the present disclosure, updating a first weight corresponding to a preferred podcast type based on the playback duration of podcast information corresponding to the preferred podcast type to obtain second preferred type data includes: In response to the cumulative playing time of the podcast information corresponding to the preferred podcast type being less than a first preset time within a preset number of playing times, lowering the first weight based on a first adjustment step; In response to the cumulative playing time of the podcast information corresponding to the preferred podcast type being greater than a second preset time within the preset number of playing times, increasing the first weight based on the first adjustment step; The first preset time length is shorter than the second preset time length.

[0060] In this embodiment, the preset number of playbacks is used to limit a range of statistical playback durations, ensuring that the duration is considered on the basis of a relatively uniform and reasonable listening frequency, and avoiding the accuracy of the judgment of the user's interest level being affected by a large difference in the number of listenings. The first preset duration and the second preset duration are used to determine the specific duration limits of the user's interest in the preferred podcast type.

[0061] When, within the preset number of playbacks, the cumulative playback time of the podcast information corresponding to the preferred podcast type is less than the first preset time, it indicates that within the set statistical period, the total time that the user listened to this type of preferred podcast was relatively short.

[0062] For example, the preset number of plays is set to 10 times, and the first preset duration is set to 30 minutes. If a preferred podcast type (such as "food recommendations") is listened to 10 times by the user, the cumulative playback time is only 20 minutes, which means that the user has invested less time in podcast content such as "food recommendations", and it is likely that the interest is gradually waning. Based on this situation, this embodiment can lower the first weight corresponding to the preferred podcast type according to the first adjustment step. For example, the original weight is 0.6, and it is reduced by a certain value (assuming a reduction of 0.1) through the first adjustment step to 0.5, so as to reflect its decreased importance in the user's interest, and its recommendation ratio will also be adjusted accordingly when customizing the podcast information later.

[0063] If within the preset number of playbacks, the cumulative playback time of the podcast information corresponding to the preferred podcast type is greater than the second preset time, it indicates that within the same statistical period, the user has listened to the preferred podcast type for a longer total time.

[0064] For example, if the preset number of plays is 10 and the second preset duration is 60 minutes, if the cumulative play duration corresponding to the preferred podcast type of "Technology Review" reaches 70 minutes, it means that the user is very interested in content such as "Technology Review" and spends a lot of time listening to it. At this time, this embodiment can increase the first weight corresponding to the preferred podcast type according to the first adjustment step, for example, from the original 0.5 to 0.6, so that it can be more prominent in the subsequent customization of podcast information, obtain more recommendation opportunities, and better meet the user's current strong interest needs.

[0065] From the above, it can be concluded that this embodiment sets a mechanism for dynamically adjusting weights based on the comparison between playback duration and preset conditions, so that this embodiment can more accurately and timely capture changes in users' interest in different preferred podcast types, and then reasonably adjust the recommendation strategy, so that the formulation of podcast information always fits the user's real listening preferences and interests, thereby increasing the probability of users obtaining satisfactory podcast content and optimizing the user experience.

[0066] In one embodiment of the present disclosure, updating a second weight corresponding to a random podcast type based on the playback duration of podcast information corresponding to the random podcast type to obtain second random type data includes: In response to the average single play time of the podcast information corresponding to the random podcast type being less than a third preset time within the preset time interval, and the number of plays being less than a first preset number, lowering the second weight based on the second adjustment step; In response to the average single play time of the podcast information corresponding to the random podcast type being greater than a fourth preset time within the preset time interval, and the number of plays being greater than a second preset number, increasing the second weight based on the second adjustment step; Among them, the third preset time length is smaller than the fourth preset time length, and the first preset number of times is smaller than the second preset number of times.

[0067] In this embodiment, the preset duration interval is a specific time period defined for statistical analysis, which aims to focus on the user's listening performance of random podcast types within a relatively stable period, so that the data can more accurately reflect the current situation. The third preset duration and the fourth preset duration are the boundary standards for measuring the duration of a single play, and the first preset number and the second preset number are used to determine the number of play times.

[0068] In this embodiment, for the random podcast type, since its initial weight is low, the role of the random podcast type is to expand the user's potential interests. If the random podcast type is processed by updating the first weight of the preferred podcast type, it is difficult to accurately capture its single listening depth and accidental discovery value based only on the cumulative playback time and the preset number of playbacks. It may misjudge the user's interest in exploring new areas, causing potential random podcasts to be misplaced or ignored, and unable to effectively tap into the user's potential preferences. Therefore, this embodiment uses a comprehensive judgment method of average single playback time combined with the number of playbacks to update the second weight corresponding to the random podcast type.

[0069] If the average single play time is very short and the number of plays is also very small, for example, the preset time interval is set to one week, the third preset time is set to 5 minutes, and the first preset number is set to 3 times, when a random podcast type (such as "Introduction to Emerging Sports") has an average single play time of only 3 minutes and is only played 2 times in a week, it means that the user has almost no interest in this type of random recommended content. The second adjustment step is used to lower its weight to further reduce its influence in the recommendation. On the contrary, if the average single play time is long and the number of plays is large, such as the fourth preset time is set to 10 minutes, and the second preset number is set to 8 times, when the random podcast type "Interesting Psychology Knowledge" has an average single play time of 12 minutes and is played 10 times in a week, it means that the user has a high interest in it. Based on the second adjustment step, the weight is increased so that it can have more opportunities to be displayed in random recommendations and gradually move closer to the preferred podcast type, thereby more reasonably exploring and strengthening the user's potential interest in the random podcast type on the basis of ensuring that it does not interfere with the main preferred recommendation, making the weight update of the random podcast type more scientific and flexible.

[0070] In this embodiment, since the initial weight of the random podcast type is low, the second adjustment step should be relatively small. For example, if the initial weight is between 0.1-0.3, the second adjustment step can be set to 0.05-0.1. This can avoid large fluctuations in weights due to a small amount of data changes, making the weight update process more stable.

[0071] For example, the initial weight of a random podcast type is 0.15. When the weight needs to be lowered based on the playback duration and number of plays, if the step size is set to 0.05, the weight will be reduced to 0.1 at most, which will not cause this random podcast type to be completely ignored in the recommendation.

[0072] From the above, it can be concluded that this embodiment makes a reasonable adjustment of the weight by comprehensively considering the average single play time and the number of plays within a specific time interval. When the time is short and the number of plays is small, the weight is lowered to reduce invalid recommendations; when the time is long and the number of plays is large, the weight is increased to explore potential interests, make random recommendations more suitable for users, and optimize the podcast customization experience.

[0073] Corresponding to the podcast information customization method of the above embodiment, Figure 2 This is a structural block diagram of a podcast information customization device provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 2 The podcast information customization device 20 includes: an information customization setting module 21, a weight distribution module 22, and an information customization module 23.

[0074] The information customization setting module 21 is used to generate a random podcast type based on multiple preferred podcast types determined by the user, where the preferred podcast type is the preferred type set by the user when using the podcast for the first time; A weight assignment module 22, configured to assign a first weight to a plurality of preferred podcast types to obtain first preferred type data, and assign a second weight to a random podcast type to obtain first random type data, wherein the first weight is greater than the first weight; The information customization module 23 is used to customize podcast information for the user based on the first preference type data and the first random type data.

[0075] In one embodiment of the present disclosure, a weight updating module; the weight updating module is specifically used for: updating a first weight corresponding to the preferred podcast type based on the playback duration of the podcast information corresponding to the preferred podcast type to obtain second preferred type data; Based on the playback duration of the podcast information corresponding to the random podcast type, a second weight corresponding to the random podcast type is updated to obtain second random type data; The podcast information is customized for the user based on the second preference type data and the second random type data.

[0076] In one embodiment of the present disclosure, the weight updating module is further configured to: generating a preference data set based on the second preference type data, and generating a random data set based on the second random type data; The preference data set is updated based on the first weight corresponding to the second preference type data, and the random data set is updated based on the second weight corresponding to the second random type data.

[0077] In one embodiment of the present disclosure, the weight updating module is further configured to: In response to a first weight corresponding to the second preference type data in the preference data set being less than a first preset weight, storing the corresponding preference type data in the preference data set into the random data set; In response to a second weight corresponding to the second random type data in the random data set being greater than or equal to the first preset weight, the corresponding random type data in the random data set is stored in the preference data set.

[0078] In one embodiment of the present disclosure, the weight updating module is further configured to: In response to the second weight corresponding to the random type data in the random data set being less than the second preset weight, deleting the corresponding random type data in the random data set; The second preset weight is smaller than the first preset weight.

[0079] In one embodiment of the present disclosure, the weight updating module is further configured to: In response to the cumulative playing time of the podcast information corresponding to the preferred podcast type being less than a first preset time within a preset number of playing times, increasing the first weight based on a first adjustment step; In response to the cumulative playing time of the podcast information corresponding to the preferred podcast type being greater than a second preset time within the preset number of playing times, lowering the first weight based on the first adjustment step; Among them, the first preset time length is shorter than the second preset time length.

[0080] In one embodiment of the present disclosure, the weight updating module is further configured to: In response to the average single play time of the podcast information corresponding to the random podcast type being less than a third preset time within the preset time interval, and the number of plays being less than a first preset number, lowering the second weight based on the second adjustment step; In response to the average single play time of the podcast information corresponding to the random podcast type being greater than a fourth preset time within the preset time interval, and the number of plays being greater than a second preset number, increasing the second weight based on the second adjustment step; Among them, the third preset time length is smaller than the fourth preset time length, and the first preset number of times is smaller than the second preset number of times.

[0081] See also Figure 3 , Figure 3 This is a schematic block diagram of a smart device provided by an embodiment of the present disclosure. Figure 3 The smart device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of each module in the above-mentioned device embodiments, such as Figure 2 The functions of modules 21 to 23 are shown.

[0082] It should be understood that in the embodiment of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0083] The input device 302 may include a touch panel, a fingerprint collection sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 303 may include a display (LCD, etc.), a speaker, etc.

[0084] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0085] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure may execute the implementation methods described in the first and second embodiments of the podcast information customization method provided in the embodiments of the present disclosure, and may also execute the implementation methods of the smart device described in the embodiments of the present disclosure, which will not be repeated here.

[0086] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0087] The computer-readable storage medium may be an internal storage unit of the smart device of any of the aforementioned embodiments, such as a hard disk or memory of the smart device. The computer-readable storage medium may also be an external storage device of the smart device, such as a plug-in hard disk equipped on the smart device, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the computer-readable storage medium may also include both an internal storage unit of the smart device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the smart device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0088] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0089] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the smart devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0090] In the several embodiments provided in the present application, it should be understood that the disclosed intelligent devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.

[0091] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0092] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0093] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A podcast information customization method, characterized in that: include: generating a random podcast type based on a plurality of preferred podcast types determined by a user, wherein the preferred podcast type is a preferred type set by the user when using a podcast for the first time; Assigning a first weight to the plurality of preferred podcast types to obtain first preferred type data, and assigning a second weight to the random podcast type to obtain first random type data, wherein the first weight is greater than the second weight; Podcast information is customized for the user based on the first preference type data and the first random type data.

2. The podcast information customization method according to claim 1, characterized in that: Also includes: updating a first weight corresponding to the preferred podcast type based on the playback duration of the podcast information corresponding to the preferred podcast type to obtain second preferred type data; Based on the playback duration of the podcast information corresponding to the random podcast type, a second weight corresponding to the random podcast type is updated to obtain second random type data; Podcast information is customized for the user based on the second preference type data and the second random type data.

3. The podcast information customization method according to claim 2, characterized in that: Before customizing podcast information for the user based on the second preference type data and the second random type data, the method further includes: generating a preference data set based on the second preference type data, and generating a random data set based on the second random type data; The preference data set is updated based on a first weight corresponding to the second preference type data, and the random data set is updated based on a second weight corresponding to the second random type data.

4. The podcast information customization method according to claim 3, characterized in that: The updating of the preference data set based on the first weight corresponding to the second preference type data, and the updating of the random data set based on the second weight corresponding to the second random type data, comprises: In response to a first weight corresponding to the second preference type data in the preference data set being less than a first preset weight, storing the corresponding preference type data in the preference data set into the random data set; In response to a second weight corresponding to the second random type data in the random data set being greater than or equal to the first preset weight, the corresponding random type data in the random data set is stored in the preference data set.

5. The podcast information customization method according to claim 4, characterized in that: Also includes: In response to the second weight corresponding to the random type data in the random data set being less than the second preset weight, deleting the corresponding random type data in the random data set; The second preset weight is smaller than the first preset weight.

6. The podcast information customization method according to claim 2, characterized in that: The updating of the first weight corresponding to the preferred podcast type based on the playback duration of the podcast information corresponding to the preferred podcast type to obtain second preferred type data includes: In response to the cumulative playing time of the podcast information corresponding to the preferred podcast type being less than a first preset time within a preset number of playing times, increasing the first weight based on a first adjustment step; In response to the cumulative playing time of the podcast information corresponding to the preferred podcast type being greater than a second preset time within a preset number of playing times, lowering the first weight based on a first adjustment step; Among them, the first preset time length is shorter than the second preset time length.

7. The podcast information customization method according to claim 2, characterized in that: The updating of the second weight corresponding to the random podcast type based on the playback duration of the podcast information corresponding to the random podcast type to obtain the second random type data includes: In response to the average single play time of the podcast information corresponding to the random podcast type being less than a third preset time within the preset time interval, and the number of plays being less than a first preset number, lowering the second weight based on the second adjustment step; In response to the average single play time of the podcast information corresponding to the random podcast type being greater than a fourth preset time within the preset time interval, and the number of plays being greater than a second preset number, increasing the second weight based on the second adjustment step; Among them, the third preset time length is smaller than the fourth preset time length, and the first preset number of times is smaller than the second preset number of times.

8. A podcast information customization device, characterized in that: include: An information customization setting module, used to generate a random podcast type based on a plurality of preferred podcast types determined by a user, wherein the preferred podcast type is a preferred type set by the user when using a podcast for the first time; A weight assignment module, configured to assign a first weight to the plurality of preferred podcast types to obtain first preferred type data, and assign a second weight to the random podcast type to obtain first random type data, wherein the first weight is greater than the second weight; An information customization module is used to customize podcast information for users based on the first preference type data and the first random type data.

9. An intelligent device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images