Method, device, server and medium for determining anchor life cycle stage

Abstract

The present disclosure relates to a method, device, server, and medium for determining a host's lifecycle stage. The method comprises: obtaining live broadcast data of a host; classifying the live broadcast data to obtain multiple types of live broadcast data, wherein each type of live broadcast data is used to determine the lifecycle stage corresponding to the type; and determining the host's lifecycle stage based on the multiple types of live broadcast data. The method for determining a host's lifecycle stage provided by the present disclosure can determine the host's lifecycle stage, so that targeted measures can be taken against the host based on the host's lifecycle stage.

Description

Technical Field

[0001] The present disclosure relates to the field of computers, and in particular to a method, device, server, and medium for determining a host's life cycle stage. Background Art

[0002] With the rapid development of electronic technology, more and more online entertainment can be carried out through electronic devices. Among them, live streaming is an emerging online entertainment method. Anchors can use electronic devices to interact with viewers on the live streaming platform.

[0003] To better analyze live streamers, it's necessary to determine their lifecycle. However, there's no solution for determining this in the related art. While related art can determine the lifecycle of consumers, this is based on consumption frequency and amount spent. The lifecycle stages of live streamers and consumers follow different patterns, making the same solution for determining the lifecycle of consumers inapplicable to the lifecycle of live streamers. Summary of the Invention

[0004] The present disclosure provides a method, device, server, and medium for determining the lifecycle stage of a streamer, to at least address the problem in related technologies of lacking a solution for determining the lifecycle stage of a streamer. The technical solutions of the present disclosure are as follows:

[0005] According to a first aspect of an embodiment of the present disclosure, a method for determining a host life cycle stage is provided, comprising:

[0006] Get the live broadcast data of the anchor;

[0007] Classifying the live broadcast data to obtain multiple types of live broadcast data, wherein each type of live broadcast data is used to determine a life cycle stage corresponding to the type;

[0008] Determine the life cycle stage of the anchor based on the multiple types of live broadcast data.

[0009] In one or more embodiments of the present disclosure, determining the anchor's life cycle stage based on the multiple types of live broadcast data includes:

[0010] Determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type, wherein the first type is any type among the multiple types;

[0011] When it is determined that the anchor is not in the life cycle stage corresponding to the first type, determine whether the anchor is in the life cycle stage corresponding to the second type based on the live broadcast data of the second type among the multiple types, wherein the second type is a type among the multiple types other than the first type.

[0012] In one or more embodiments of the present disclosure, the first type of live broadcast data includes the duration of the anchor's off-broadcasting;

[0013] The determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type includes:

[0014] If the anchor's off-broadcasting duration is between the first off-broadcasting duration threshold and the second off-broadcasting duration threshold, it is determined that the anchor is in a dormant period;

[0015] If the anchor's off-broadcasting duration is greater than the second off-broadcasting duration threshold, it is determined that the anchor is in a loss period.

[0016] In one or more embodiments of the present disclosure, the method of determining whether the anchor is in a life cycle stage corresponding to a first type of live broadcast data among the multiple types based on the first type of live broadcast data further includes:

[0017] Obtaining a correspondence between the broadcaster's off-broadcast duration and the return rate, where the return rate represents: the ratio of the number of broadcasters who resume live streaming within a predetermined time period among the M sample broadcasters who have reached the off-broadcast duration to the total number of the M sample broadcasters, where M is an integer greater than 1;

[0018] According to the corresponding relationship, obtaining the suspension duration corresponding to the first preset reflux rate and the suspension duration corresponding to the second preset reflux rate;

[0019] The suspension duration corresponding to the first preset reflux rate is determined as the first suspension duration threshold, and the suspension duration corresponding to the second preset reflux rate is determined as the second suspension duration threshold.

[0020] In one or more embodiments of the present disclosure, the first type of live broadcast data includes a live broadcast data sequence of the anchor;

[0021] The determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type includes:

[0022] Calculating a coefficient of variation of the live data sequence based on values ​​in the live data sequence, wherein the coefficient of variation represents a degree of dispersion of the live data sequence;

[0023] When the coefficient of variation is within a preset range, it is determined that the anchor is in the mature stage.

[0024] In one or more embodiments of the present disclosure, the method of determining whether the anchor is in a life cycle stage corresponding to a first type of live broadcast data among the multiple types based on the first type of live broadcast data further includes:

[0025] Obtaining a correspondence between the repetition rate of sample hosts and the statistical period, wherein the repetition rate represents: after determining N sample hosts in the mature stage based on the live broadcast data sequence within the statistical period, the proportion of the number of sample hosts that continue to be in the mature stage within a preset time period to the total number of the N sample hosts, where N is an integer greater than 1;

[0026] According to the corresponding relationship, obtaining a target period corresponding to a predetermined repetition rate;

[0027] The classifying of the live broadcast data to obtain multiple types of live broadcast data includes:

[0028] According to the target period, the live broadcast data sequence of the anchor is extracted from the live broadcast data, wherein a value of the live broadcast data sequence is obtained based on statistics of the live broadcast data within the target period.

[0029] In one or more embodiments of the present disclosure, before obtaining the live broadcast data of the anchor, the method further includes:

[0030] Obtain data on multiple live broadcast indicators of sample anchors;

[0031] Calculate the coefficient of variation of each of the live broadcast indicators based on the data of each of the live broadcast indicators;

[0032] Obtaining a target live broadcast indicator with the smallest coefficient of variation among the multiple live broadcast indicators;

[0033] The method of obtaining the live broadcast data of the anchor includes:

[0034] According to the target live broadcast indicator, data of the target live broadcast indicator of the anchor is obtained.

[0035] In one or more embodiments of the present disclosure, the first type of live broadcast data includes the host's target live broadcast time point, where the target live broadcast time point is the time point when the host performs the first live broadcast, or the time point when the host performs the live broadcast during the churn period;

[0036] The determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type includes:

[0037] According to the target live broadcast time point, it is determined that the anchor is in the emerging stage in a first time period after the target live broadcast time point, and it is determined that the anchor is in the growing stage in a second time period after the target live broadcast time point.

[0038] The duration of the first time period is shorter than the duration of the second time period.

[0039] In one or more embodiments of the present disclosure, the first type of live broadcast data includes a first broadcast frequency, a second broadcast frequency, and a continuous non-live broadcast duration of the anchor, wherein the first broadcast frequency is the broadcast frequency of the anchor in a recent time period, and the second broadcast frequency is the broadcast frequency of the anchor in another time period before the time period;

[0040] The determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type includes:

[0041] When the first broadcast frequency is less than the second broadcast frequency, or the number of consecutive days without live broadcast is greater than a preset threshold, it is determined that the anchor is in a decline period;

[0042] When the first broadcast frequency is not less than the second broadcast frequency, and the number of consecutive days without live broadcast is not greater than a preset day threshold, it is determined that the anchor is in the growth stage.

[0043] According to a second aspect of an embodiment of the present disclosure, a device for determining a host life cycle stage is provided, including:

[0044] A first acquisition module is configured to acquire live broadcast data of the anchor;

[0045] a classification module configured to classify the live broadcast data to obtain multiple types of live broadcast data, wherein each type of live broadcast data is used to determine a life cycle stage corresponding to the type;

[0046] The first determination module is configured to determine the life cycle stage of the anchor based on the multiple types of live broadcast data.

[0047] In one or more embodiments of the present disclosure, the first determining module includes:

[0048] A first determining unit is configured to determine, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type, wherein the first type is any type among the multiple types;

[0049] The second determination unit is configured to determine whether the anchor is in the life cycle stage corresponding to the first type based on the live broadcast data of the second type among the multiple types when it is determined that the anchor is not in the life cycle stage corresponding to the first type, wherein the second type is a type among the multiple types other than the first type.

[0050] In one or more embodiments of the present disclosure, the first type of live broadcast data includes the duration of the anchor's off-broadcasting;

[0051] The first determining unit is specifically configured to:

[0052] If the anchor's off-broadcasting duration is between the first off-broadcasting duration threshold and the second off-broadcasting duration threshold, it is determined that the anchor is in a dormant period;

[0053] If the anchor's off-broadcasting duration is greater than the second off-broadcasting duration threshold, it is determined that the anchor is in a loss period.

[0054] In one or more embodiments of the present disclosure, the apparatus further includes:

[0055] The second acquisition module is configured to obtain a correspondence between the broadcaster's off-broadcast duration and the reflow rate, wherein the reflow rate represents: the ratio of the number of broadcasters who resume live broadcasting within a predetermined time period among the M sample broadcasters who have reached the off-broadcast duration to the total number of the M sample broadcasters, where M is an integer greater than 1;

[0056] A third acquisition module is configured to acquire, according to the corresponding relationship, the suspension duration corresponding to the first preset reflux rate and the suspension duration corresponding to the second preset reflux rate;

[0057] The second determining module is configured to determine the suspension duration corresponding to the first preset reflux rate as the first suspension duration threshold, and determine the suspension duration corresponding to the second preset reflux rate as the second suspension duration threshold.

[0058] In one or more embodiments of the present disclosure, the first type of live broadcast data includes a live broadcast data sequence of the anchor;

[0059] The first determining unit includes:

[0060] a first calculation subunit configured to calculate a coefficient of variation of the live data sequence based on values ​​in the live data sequence, wherein the coefficient of variation represents a degree of dispersion of the live data sequence;

[0061] The first determining subunit is configured to determine that the anchor is in the mature stage when the coefficient of variation is within a preset range.

[0062] In one or more embodiments of the present disclosure, the apparatus further includes:

[0063] A fourth acquisition module is configured to obtain a correspondence between a repetition rate of sample anchors and a statistical period, wherein the repetition rate represents: after N sample anchors in the mature stage are determined based on the live broadcast data sequence within the statistical period, the proportion of the number of sample anchors that continue to be in the mature stage within a preset time period to the total number of the N sample anchors, where N is an integer greater than 1;

[0064] A fifth acquisition module is configured to acquire a target period corresponding to the predetermined repetition rate according to the corresponding relationship;

[0065] Wherein, the classification module is specifically configured as follows:

[0066] According to the target period, the live broadcast data sequence of the anchor is extracted from the live broadcast data, wherein a value of the live broadcast data sequence is obtained based on statistics of the live broadcast data within the target period.

[0067] In one or more embodiments of the present disclosure, the apparatus further includes:

[0068] a sixth acquisition module, configured to acquire data of multiple live broadcast indicators of the sample anchor;

[0069] A calculation module is configured to calculate the coefficient of variation of each of the live broadcast indicators based on the data of each of the live broadcast indicators;

[0070] A seventh acquisition module is configured to acquire a target live broadcast indicator with the smallest coefficient of variation among the multiple live broadcast indicators;

[0071] The first acquisition module is specifically configured to: acquire data of the target live broadcast indicator of the anchor according to the target live broadcast indicator.

[0072] In one or more embodiments of the present disclosure, the first type of live broadcast data includes the host's target live broadcast time point, where the target live broadcast time point is the time point when the host performs the first live broadcast, or the time point when the host performs the live broadcast during the churn period;

[0073] The first determining unit is configured to:

[0074] According to the target live broadcast time point, it is determined that the anchor is in the emerging stage in a first time period after the target live broadcast time point, and it is determined that the anchor is in the growing stage in a second time period after the target live broadcast time point.

[0075] The duration of the first time period is shorter than the duration of the second time period.

[0076] In one or more embodiments of the present disclosure, the first type of live broadcast data includes a first broadcast frequency, a second broadcast frequency, and a continuous non-live broadcast duration of the anchor, wherein the first broadcast frequency is the broadcast frequency of the anchor in a recent time period, and the second broadcast frequency is the broadcast frequency of the anchor in another time period before the time period;

[0077] The first determining unit is configured to:

[0078] When the first broadcast frequency is less than the second broadcast frequency, or the number of consecutive days without live broadcast is greater than a preset threshold, it is determined that the anchor is in a decline period;

[0079] When the first broadcast frequency is not less than the second broadcast frequency, and the number of consecutive days without live broadcast is not greater than a preset day threshold, it is determined that the anchor is in the growth stage.

[0080] According to a third aspect of an embodiment of the present disclosure, a server is provided, including:

[0081] processor;

[0082] a memory for storing instructions executable by the processor;

[0083] Among them, the processor is configured to execute the instructions to implement the method for determining the anchor life cycle stage provided in the first aspect.

[0084] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When the instructions in the computer-readable storage medium are executed by a processor of a server, the server is enabled to execute the method for determining the anchor life cycle stage provided in the first aspect.

[0085] According to a fifth aspect of an embodiment of the present disclosure, a method is provided including a computer program / instruction, which, when executed by a processor, implements the method for determining the anchor life cycle stage provided in the first aspect.

[0086] The technical solutions provided by the embodiments of the present disclosure bring at least the following beneficial effects:

[0087] First, the live broadcast data of the host is obtained. Since the live broadcast data includes multiple types of live broadcast data, and different types of live broadcast data are used to determine different life cycle stages of the host, the live broadcast data is classified to obtain multiple types of live broadcast data. Then, based on the multiple types of live broadcast data, the host's life cycle stage is determined. This allows the host's life cycle stage to be determined, allowing targeted measures to be taken based on the host's life cycle stage.

[0088] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.

[0090] Figure 1 This is a schematic diagram showing the relationship between various stages of a host's life cycle and a description of the host's behavior according to an exemplary embodiment.

[0091] Figure 2 The figure is a flowchart of a method for determining a host's life cycle stage according to an exemplary embodiment.

[0092] Figure 3 The figure is a schematic diagram showing the relationship between the return rate and the number of days of suspension according to an exemplary embodiment.

[0093] Figure 4 is a schematic diagram showing stationarity detection results obtained by using different algorithms according to an exemplary embodiment.

[0094] Figure 5 The figure is a schematic diagram showing the relationship between the statistical period and the repetition rate according to an exemplary embodiment.

[0095] Figure 6 The figure is a schematic diagram of a curve showing the relationship between the statistical period and the churn rate according to an exemplary embodiment.

[0096] Figure 7 The figure is a flowchart of a method for determining a host life cycle stage according to another exemplary embodiment.

[0097] Figure 8 The present invention is a flowchart illustrating a process of constructing a model for determining the life cycle stages of an anchor according to an exemplary embodiment.

[0098] Figure 9The figure is a schematic diagram of a framework of a model for determining the life cycle stages of a host according to an exemplary embodiment.

[0099] Figure 10 The figure is a schematic structural diagram of a device for determining a host life cycle stage according to an exemplary embodiment.

[0100] Figure 11 The figure is a structural block diagram of a server according to an exemplary embodiment. DETAILED DESCRIPTION

[0101] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0102] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0103] The life cycle stage determination method in related technologies focuses on consumers. For example, in the e-commerce field, the user life cycle stage is mainly divided into the user's life cycle stage by consumer behaviors such as purchase frequency and purchase amount, combined with thresholds obtained by data analysis.

[0104] However, the life cycle stage determination method in the related art cannot be applied to creators such as anchors, as the starting points of the two are different. In the method for determining the life cycle stage of consumers, the division of the life cycle mainly considers whether the cumulative consumption frequency and consumption amount within a certain statistical time period reach a certain threshold, thereby obtaining the user's life cycle stage. For a platform, it is unreasonable to judge whether the anchor is mature, growing or declining based on the cumulative behavior over a period of time. Because the anchor's behavior is cyclical, mature anchors have stable broadcasting behavior, and anchors with low frequency but stable cycles are also mature anchors. Therefore, the life cycle stage determination method in the related art cannot be applied to creators such as anchors, and there is no solution to determine the anchor's life cycle stage in the related art.

[0105] In order to solve the problem that there is no solution for determining the life cycle of an anchor in the related art, the present disclosure provides a method for determining the life cycle stage of an anchor. Before explaining the method for determining the life cycle stage of an anchor provided by the present disclosure, the various stages of the anchor's life cycle are first explained.

[0106] The various stages of the anchor life cycle can be divided into the following stages based on business understanding: introduction period, new period, growth period, maturity period, decline period, dormancy period, and loss period. The relationship between the various stages of the anchor life cycle and the description of anchor behavior can be found in Figure 1 ,from Figure 1 It can be seen that the different stages of the anchor's life cycle are determined based on the anchor's different live broadcast data.

[0107] Figure 2 FIG. 1 is a flow chart showing a method for determining a host's life cycle stage according to an exemplary embodiment. Figure 2 As shown, the method for determining the anchor life cycle stage includes:

[0108] S102, obtaining the live broadcast data of the anchor.

[0109] Among them, the live broadcast data of the anchor includes at least one of the following: the anchor's off-broadcast duration, the anchor's live broadcast data sequence (such as the sequence of the anchor's live broadcast days in multiple time periods), the time point when the anchor first broadcasts live, the time point when the anchor broadcasts live during the attrition period, and the broadcast frequency.

[0110] As an example, the live broadcast data of each anchor on the live broadcast application can be stored in a database. Then, in S102, the live broadcast data of the anchor can be obtained from the database according to the account of the anchor in the life cycle stage to be determined.

[0111] Methods for determining the anchor life cycle stage also include:

[0112] S104: Classify the live broadcast data to obtain multiple types of live broadcast data, wherein each type of live broadcast data is used to determine a life cycle stage corresponding to the type.

[0113] For example, the length of time the anchor is off-broadcast is regarded as one type of live broadcast data, the number of days the anchor broadcasts live is regarded as another type of live broadcast data, the time when the anchor broadcasts for the first time and the time when the anchor broadcasts live during the loss period are regarded as another type of live broadcast data, and the frequency of the anchor's broadcast is regarded as yet another type of live broadcast data.

[0114] Methods for determining the anchor life cycle stage also include:

[0115] S106: Determine the anchor's life cycle stage based on multiple types of live broadcast data.

[0116] Among them, in S106, first determine whether the anchor is in the life cycle stage corresponding to a type based on live broadcast data of the type; if the anchor is not in the life cycle stage corresponding to the type, then determine the life cycle stage of the anchor based on live broadcast data of another type.

[0117] Specifically, S106 may include:

[0118] Determining, based on live broadcast data of a first type among multiple types, whether the anchor is in a life cycle stage corresponding to the first type, wherein the first type is any type among the multiple types;

[0119] When it is determined that the anchor is not in the life cycle stage corresponding to the first type, determine whether the anchor is in the life cycle stage corresponding to the second type based on the live broadcast data of the second type among multiple types, where the second type is a type among the multiple types other than the first type.

[0120] The multiple types of live broadcast data may be arranged in a predetermined order, or may be sorted according to corresponding life cycle stages. If it is determined based on the first type of live broadcast data that the host is not in the life cycle stage corresponding to the first type, the host's life cycle stage may be determined based on the second type of live broadcast data subsequent to the first type.

[0121] For example, first determine whether the anchor is in the growth stage based on the anchor's broadcast frequency. If it is determined that the anchor is not in the growth stage, then determine whether the anchor is in the mature stage based on the anchor's live broadcast data sequence, and so on, until the anchor's life cycle stage is determined.

[0122] In the disclosed embodiment, first, a host's live broadcast data is obtained. Then, because the host's live broadcast data includes multiple types of live broadcast data, and different types of live broadcast data are used to determine different lifecycle stages of the host, the live broadcast data is classified to obtain multiple types of live broadcast data. Then, based on the multiple types of live broadcast data, the host's lifecycle stage is determined. This allows the host's lifecycle stage to be determined, allowing targeted measures to be taken based on the host's lifecycle stage.

[0123] In one or more embodiments of the present disclosure, the first type of live broadcast data includes the duration of the host's off-airing period. The off-air duration includes the number of days or hours the host has been off-air. The off-air duration of the host can be used to determine whether the host is in a dormant period or a period of churn.

[0124] Determining whether the anchor is in a life cycle stage corresponding to the first type based on live broadcast data of the first type among the multiple types may specifically include:

[0125] If the anchor's off-broadcasting duration is between the first off-broadcasting duration threshold and the second off-broadcasting duration threshold, it is determined that the anchor is in a dormant period;

[0126] If the anchor's off-broadcasting duration is greater than the second off-broadcasting duration threshold, it is determined that the anchor is in the churn period.

[0127] The first and second pause duration thresholds can be determined based on the concept of returning visitors and the return rate of the anchor after the anchor stops broadcasting.

[0128] Specifically, before determining whether the anchor is in the life cycle stage corresponding to the first type according to the live broadcast data of the first type among the multiple types, the method for determining the anchor life cycle stage may further include:

[0129] Obtain the corresponding relationship between the broadcaster's off-air duration and the return rate, where the return rate represents: the ratio of the number of broadcasters who resume live streaming within the predetermined time period among the M sample broadcasters who have reached the off-air duration, to the total number of the M sample broadcasters, where M is an integer greater than 1;

[0130] According to the correspondence between the anchor's off-broadcasting duration and the reflux rate, the off-broadcasting duration corresponding to the first preset reflux rate and the off-broadcasting duration corresponding to the second preset reflux rate are obtained;

[0131] The suspension duration corresponding to the first preset reflux rate is determined as a first suspension duration threshold, and the suspension duration corresponding to the second preset reflux rate is determined as a second suspension duration threshold.

[0132] The following combination Figure 3 The following describes how to determine a first pause duration threshold and a second pause duration threshold based on a reflow rate.

[0133] Figure 3 is a diagram showing a curve relationship between a return rate and the number of days of suspension of broadcasting according to an exemplary embodiment. Figure 3 The curve relationship shown is obtained by analyzing the live broadcast behaviors of multiple sample anchors. Figure 3 It shows that the return of sample anchors with different suspension days in the next 30 days shows a downward trend. The suspension days corresponding to the return rate X1% are represented by t1: among the sample anchors who stopped broadcasting for t1 days, the number of sample anchors who broadcast live again within 30 days after t1 accounts for the proportion of the total number of sample anchors, and t1 is a positive integer.

[0134] from Figure 3It can be seen that the curve has obvious inflection points at the first preset reflux rate X1% and the second preset reflux rate X2%, and the reflux rates are at a relatively low and stable level after the two inflection points. Figure 3 The curve shown above shows the number of days of suspension corresponding to the first preset return rate X1% (i.e., t1 days) and the number of days of suspension corresponding to the second preset return rate X2% (i.e., t2 days). If the anchor's suspension days are between t1 and t2, it can be determined that the anchor is in a dormant period. If the anchor's suspension days are greater than t2, it can be determined that the anchor is in a churn period.

[0135] In one or more embodiments of the present disclosure, the first type of live broadcast data includes a live broadcast data sequence of the anchor. For example, the live broadcast data sequence is a broadcast day sequence, and the values ​​in the broadcast day sequence are the broadcast days of the anchor in each time period.

[0136] A streamer's live streaming data series can be used to determine whether they are in the mature stage. Since the mature stage refers to streamers with stable live streaming, while streamers in the emerging, growth, and decline stages are all inactive, choosing the right method to determine whether a streamer is in the mature stage is fundamental to analyzing other life cycle stages.

[0137] The stationarity detection of time series is usually implemented using the Augmented Dickey-Fuller (ADF) test. This stationarity detection algorithm mainly requires whether the sequence after differentiation conforms to white noise, which will judge some anchor behaviors that are high and low at the beginning of the broadcast as stationary behavior. In addition, other common stationarity detection algorithms also have certain problems and are not suitable for current application scenarios. Among them, other common stationarity detection algorithms include the Local Outlier Factor (LOF) algorithm and the 2Sigma detection algorithm. Therefore, it is particularly important to choose a suitable stationarity detection algorithm to process the live broadcast data sequence to determine whether the anchor is in the mature stage.

[0138] Below through Figure 4 To illustrate how to choose a suitable stability detection algorithm to determine whether the anchor is in the mature stage. Figure 4 Each broadcast day sequence in is the broadcast day sequence of the anchor for 8 weeks, and the value in the broadcast day sequence is the broadcast day counted every other week. Figure 4 The sequence of broadcast days is [7, 7, 7, 7, 7, 7, 7], the first "7" in the sequence means that the host will broadcast for 7 days in the first week, the second "7" means that the host will broadcast for 7 days in the second week, and so on.

[0139] Figure 4The results of the stationary detection of the sequence of broadcast days using different detection algorithms are shown. Figure 4 As can be seen, the ADF stationarity test and 2Sigma test results for impulsive and spontaneous streamers do not meet the definition of maturity. The LOF anomaly test is more stringent for occasional streamers. Relatively speaking, the coefficient of variation test, while having some misjudgments for spontaneous streamers, performs relatively well for other common streamer types. Therefore, the coefficient of variation is chosen as the detection method for streamer behavior stability, rather than the commonly used ADF, 2Sigma, or LOF tests.

[0140] Based on the above analysis, determining whether the anchor is in the life cycle stage corresponding to the first type according to the live broadcast data of the first type among the multiple types may specifically include:

[0141] Calculate the coefficient of variation of the live data sequence based on the values ​​in the live data sequence, where the coefficient of variation represents the degree of dispersion of the live data sequence;

[0142] If the coefficient of variation is within a preset range (for example, less than a predetermined threshold), the variation in the live stream data sequence is relatively small, indicating that the streamer has relatively stable broadcasting behavior, and is therefore considered mature. The formula for calculating the coefficient of variation is: coefficient of variation = standard deviation / mean.

[0143] When determining whether a host is in the mature stage, it is necessary to use the live broadcast data sequence of the host within a period of time. If the period of the live broadcast data sequence is too short, the host's life cycle stage may frequently jump. Conversely, if the period is too long, the migration of the host's life cycle stage may not be detected in time, which is not conducive to the formulation of operation strategies.

[0144] In order to scientifically and reasonably determine the statistical period of the live broadcast data sequence, in one or more embodiments of the present disclosure, before S102, the method for determining the anchor life cycle stage may further include:

[0145] Obtain the correspondence between the repetition rate of sample anchors and the statistical period, where the repetition rate represents: after determining N sample anchors in the mature stage based on the live broadcast data sequence within the statistical period, the proportion of the number of sample anchors that continue to be in the mature stage within a preset time period to the total number of N sample anchors, where N is an integer greater than 1;

[0146] According to the correspondence between the repetition rate of the sample anchor and the statistical period, the target period corresponding to the predetermined repetition rate is obtained.

[0147] The method of classifying the live broadcast data to obtain multiple types of live broadcast data includes extracting the live broadcast data sequence of the anchor from the live broadcast data according to the target period, wherein a value of the live broadcast data sequence is obtained based on statistics of the live broadcast data within a target period.

[0148] The following combination Figure 5 The embodiments of the present disclosure are exemplified.

[0149] Figure 5 The figure is a schematic diagram showing the relationship between the statistical period and the repetition rate according to an exemplary embodiment. Figure 5 The curve relationship shown is obtained by analyzing the live broadcast behaviors of multiple sample anchors. Specifically, Figure 5 The figure shows the proportion curve of sample anchors in the mature stage determined according to different statistical periods, and the proportion curve of sample anchors still in the mature stage after 14 days.

[0150] from Figure 5 It can be seen that the repetition rate of the sample anchors steadily increases as the statistical period increases. Using the sequence of live broadcast days in the last 8 weeks, we can calculate the proportion of sample anchors who are still mature anchors after 14 days in the mature stage. The repetition rate after 8 weeks is in a stage of slow growth, so 8 weeks can be selected as the target period.

[0151] In the disclosed embodiment, a target period is determined based on the correspondence between the repetition rate of sample livestreamers and the statistical period, and a livestream data sequence is obtained based on the target period. This prevents frequent transitions between identified livestreamer lifecycle stages due to a short statistical period and allows for timely detection of transitions, facilitating operational strategy development.

[0152] Because live streaming involves numerous unstable factors, such as the duration of a streamer's broadcast and the average concurrent users (ACU), which fluctuate widely, it's necessary to select relatively stable indicators from among multiple live streaming metrics. Generally speaking, the selected indicators need to be stable and representative. In related technologies, this approach typically relies on the staff's operational experience to select from a wide range of indicators. However, this approach relies heavily on subjective experience and is labor-intensive when there are numerous indicators.

[0153] In order to avoid relying on the subjective experience of staff when selecting indicators and to reduce manpower consumption, in one or more embodiments of the present disclosure, before S102, in one or more embodiments of the present disclosure, the method for determining the anchor life cycle stage may further include:

[0154] Obtain data on multiple live broadcast indicators of sample anchors;

[0155] Calculate the coefficient of variation of each live broadcast indicator based on the data of each live broadcast indicator;

[0156] A target live broadcast indicator with the smallest coefficient of variation is obtained from multiple live broadcast indicators. Specifically, the target live broadcast indicator may be the number of days of broadcasting.

[0157] S102 includes: obtaining data of the host's target live broadcast indicator according to the target live broadcast indicator.

[0158] The following is an example of how to obtain the target live broadcast indicator.

[0159] For example, the live broadcast data of the sample anchor within 8 weeks can be selected. The live broadcast data includes data of multiple live broadcast indicators, among which the multiple live broadcast indicators include the number of broadcast days, the number of broadcasts, the streaming time (that is, the total broadcast time of the anchor), the playback time (that is, the total time all viewers watch the live broadcast) and ACU.

[0160] According to the data of each live broadcast indicator, the coefficient of variation of each live broadcast indicator is calculated. The calculation results of the coefficient of variation of each live broadcast indicator are shown in Table 1:

[0161] Table 1

[0162] Indicator name Days of broadcast Number of broadcasts Streaming duration Playback duration Average ACU Coefficient of variation (median) 0.427 0.549 0.591 0.678 0.479

[0163] It should be noted that when calculating the coefficient of variation for broadcast days, the coefficient of variation for each sample livestreamer was first calculated. Specifically, the coefficient of variation for that sample livestreamer was calculated based on the number of broadcast days the sample livestreamer had in each of the eight weeks. After calculating the coefficient of variation for each sample livestreamer, the median of the coefficients of variation for the multiple sample livestreamers was taken to obtain the coefficient of variation for broadcast days in Table 1. Similarly, the coefficients of variation for the other livestreaming indicators in Table 1 can be calculated.

[0164] After obtaining the coefficient of variation for each live broadcast indicator in Table 1, the live broadcast indicator with the smallest coefficient of variation is selected, namely the number of days since broadcast. This allows the automatic selection of the indicator used to determine whether the anchor is in the mature stage from among the many live broadcast indicators.

[0165] In the disclosed embodiment, the coefficient of variation of each live broadcast indicator is calculated, and a target live broadcast indicator is obtained from the numerous live broadcast indicators based on the coefficient of variation. This allows the determination of whether a live broadcaster is in the mature stage based on the data from the target live broadcast indicator. This eliminates the need for human intervention in indicator selection, avoids relying on subjective experience of staff members, and reduces manual intervention. Even when selecting from a large number of indicators, no human effort is required.

[0166] In one or more embodiments of the present disclosure, the first type of live broadcast data includes the host's target live broadcast time point, which is the time point when the host performs the first live broadcast, or the time point when the host performs the live broadcast during the loss period.

[0167] Determining, based on live broadcast data of a first type among the multiple types, whether the host is in a life cycle stage corresponding to the first type may include:

[0168] According to the target live broadcast time point, it is determined that the anchor is in the newborn stage within a first time period after the target live broadcast time point, and it is determined that the anchor is in the growth stage within a second time period after the target live broadcast time point, wherein the length of the first time period is less than the length of the second time period.

[0169] The following combination Figure 6 Describe how to determine the first time period and the second time period. Figure 6 FIG. 1 is a schematic diagram of a curve showing the relationship between the statistical period and the churn rate according to an exemplary embodiment. Figure 6 The horizontal axis represents the statistical period, and the vertical axis represents the churn rate. The churn rate corresponding to the statistical period T represents the proportion of the number of sample anchors who did not broadcast live to the total number of sample anchors T days after the first broadcast of multiple sample anchors (that is, T+1 day after the first broadcast).

[0170] from Figure 6 It can be seen that the churn rate of multiple sample anchors on the first day of their debut was as high as 24%. That is to say, the number of anchors who did not broadcast live the day after the debut of multiple sample anchors accounted for as high as 24% of the total number of sample anchors. It can also be seen that the churn rate of multiple sample anchors dropped rapidly to 1% within 7 days of their debut (that is, the 8th day after the premiere). After 28 days, the anchor churn rate stabilized at around 0.5%.

[0171] Since churn rates at 1% and 0.5% are two inflection points, the first time period may be determined as 1 week, and the second time period may be determined as between 3 weeks and 4 weeks.

[0172] In other words, after a streamer's first broadcast, the first week is the streamer's start-up period, and the first three to four weeks are the streamer's growth period. Alternatively, if a streamer is live streaming during a period of churn, the first week is the streamer's start-up period, and the first three to four weeks are the streamer's growth period.

[0173] In defining the various stages of a streamer's lifecycle, we find that streamers in the growth and decline stages have similar broadcast patterns over a period of time, but their broadcast frequency exhibits opposite trends: the frequency of streamers in the growth stage increases continuously, while that of streamers in the decline stage decreases continuously. Therefore, the change in a streamer's broadcast frequency can be used to determine whether they are in the growth or decline stage.

[0174] Based on the above analysis, in one or more embodiments of the present disclosure, the first type of live broadcast data includes the first broadcast frequency, the second broadcast frequency and the continuous non-broadcast time of the anchor, wherein the first broadcast frequency is the broadcast frequency of the anchor in the most recent time period, and the second broadcast frequency is the broadcast frequency of the anchor in another time period before the time period.

[0175] As an example, the start time point of the most recent time period is the end time point of another time period, and the duration of these two time periods can be the same. For example, the first broadcast frequency is the host's broadcast frequency in the last 4 weeks, and the second broadcast frequency is the host's broadcast frequency in the previous 4 weeks (that is, the 4 weeks before the last 4 weeks). For another example, the first broadcast frequency is the host's broadcast frequency from April 7 to today (May 1), and the second broadcast frequency is the host's broadcast frequency from March 10 to April 7.

[0176] Determining, based on live broadcast data of a first type among the multiple types, whether the host is in a life cycle stage corresponding to the first type may specifically include:

[0177] When the first broadcast frequency is less than the second broadcast frequency, or the number of consecutive days without live broadcast is greater than a preset threshold, it is determined that the anchor is in a decline period;

[0178] When the first broadcast frequency is not less than the second broadcast frequency and the number of consecutive days without live broadcast is not greater than a preset day threshold, it is determined that the anchor is in the growth stage.

[0179] In the embodiment of the present disclosure, it can be determined whether the anchor is in the decline period or the growth period based on the anchor's broadcast frequency and the number of consecutive days without live broadcast.

[0180] Below through Figure 7 The method for determining the anchor life cycle stage provided by the present disclosure is further explained.

[0181] Figure 7 FIG. 1 is a flow chart showing a method for determining a host life cycle stage according to another exemplary embodiment. Figure 7 As shown, the method for determining the anchor life cycle stage includes:

[0182] S202, obtaining the live broadcast data of the anchor;

[0183] S204: Determine whether the anchor is in the introduction stage, nascent stage, mature stage, or dormant stage. If the judgment result is yes, it means that the anchor's life cycle stage has been determined and this process can be terminated. If the judgment result is no, execute S206;

[0184] S206: Determine whether the host's first broadcast frequency in the most recent week is less than the second broadcast frequency in the previous week, whether the host's first broadcast frequency in the most recent two weeks is less than the second broadcast frequency in the previous two weeks, and whether the host's first broadcast frequency in the most recent four weeks is less than the second broadcast frequency in the previous four weeks. If the judgment result is yes, execute step S212; otherwise, execute step S208.

[0185] S208: Determine whether the host has not broadcast live for three consecutive weeks. If the judgment result is yes, execute step S212; if the judgment result is no, execute step S210;

[0186] S210, determining that the anchor is in the growth stage;

[0187] S212: Determine whether the anchor is in a decline period.

[0188] A model for determining the host life cycle stage may be constructed, and the method for determining the host life cycle stage of the above embodiment may be executed based on the model for determining the host life cycle stage.

[0189] Figure 8 This is a flow chart showing a method for constructing a model for determining the life cycle stages of a host according to an exemplary embodiment. Figure 8 As shown in the figure, the model for determining the anchor life cycle stages specifically includes:

[0190] Conduct a logical analysis of the live broadcast business to obtain the logical analysis results of the live broadcast business. The logic of the live broadcast business includes: anchor registration on Kuaishou, activation of live broadcast permissions, broadcast start, sharing, promotion, loss, and return;

[0191] Based on the logical analysis of the live streaming business, we define the various stages of the anchor life cycle. These stages can include: introduction, nascent stage, growth stage, maturity stage, decline stage, dormancy stage, and churn stage.

[0192] Determine the behavior descriptions of each stage of the anchor life cycle, and continue to refer to Figure 1 , Figure 1 Shows the behavior description of each stage of the anchor life cycle;

[0193] Based on the behavioral descriptions of each stage of the anchor's life cycle, select the corresponding processing method to determine the anchor's life cycle stage and label the anchor's life cycle stage;

[0194] Analyze the anchor's data based on the anchor's life cycle stage to determine whether the anchor's life cycle stage is accurate. If the anchor's life cycle stage is inaccurate, fine-tune the anchor's life cycle stage behavior description to continuously optimize the anchor's life cycle stage behavior description;

[0195] After determining the anchor's life cycle stage, different operating strategies can be adopted for anchors in different life cycle stages. For example, increase the number of fans of anchors in the introduction period, increase the commission for anchors in the mature period, and increase the exposure of anchors in the return period, thereby promoting the conversion of anchors in the introduction period, extending the life stage of anchors in the mature period, and recalling anchors in the loss period.

[0196] Among them, when determining the anchor's life cycle stage, a determination model of the anchor's life cycle stage can be constructed based on the behavioral descriptions of each stage of the anchor's life cycle, and the determination model of the anchor's life cycle stage can be used to select the corresponding processing method to determine the anchor's life cycle stage.

[0197] Below through Figure 9 The example illustrates the framework of the model for determining the anchor life cycle stage provided by the present disclosure.

[0198] Figure 9 FIG. 1 is a schematic diagram of a framework of a model for determining the life cycle stages of a host according to an exemplary embodiment. Figure 9 As shown, in order to determine the anchor's life cycle stage, first obtain the anchor's various live broadcast data, and then obtain the anchor's broadcast authority information and the anchor's live broadcast indicator data from the anchor's various live broadcast data.

[0199] For the broadcast permission information, if the broadcast permission information indicates that the anchor has opened the live broadcast permission and has already made the first broadcast, it is determined that the anchor is already in the introduction period; for the live broadcast indicator data, the anchor's time series data is extracted from the live broadcast indicator data, and the time series data includes at least one of the following: the anchor's off-broadcast duration, the sequence of live broadcast days, the time point when the anchor first broadcasts, the time point when the anchor broadcasts during the loss period, and the anchor's broadcast frequency.

[0200] Then, periodic identification is performed, specifically identifying the first time period (such as 1 week) and the second time period (such as 3-4 weeks) used to determine whether the anchor is in the nascent stage and the growth stage, and the statistical period of the live broadcast day sequence used to determine whether the anchor is in the mature stage.

[0201] Determine the anchor's life cycle stage based on the anchor's time series data. The specific steps are as follows:

[0202] Perform new / return traffic identification: When the anchor conducts the first live broadcast or the live broadcast in the churn period, the first time period (for example, 1 week) after the time point of the anchor's first live broadcast or the time point of the live broadcast in the churn period is determined as the newborn period, and the second time period (for example, 3-4 weeks) after the time point of the anchor's first live broadcast or the time point of the live broadcast in the churn period is determined as the growth period.

[0203] Trend detection: Determine whether the anchor is in a decline or growth period based on the anchor's first broadcast frequency in the most recent third time period, the second broadcast frequency in the fourth time period, and the anchor's continuous non-live broadcast time.

[0204] Stability detection: Based on the live broadcast day sequence of the anchor, calculate the coefficient of variation of the live broadcast day sequence, and determine whether the anchor is in the mature stage based on the coefficient of variation.

[0205] Silence recognition: Determine the relationship between the anchor's off-broadcasting duration and the first and second off-broadcasting duration thresholds. If the anchor's off-broadcasting duration is between the first and second off-broadcasting duration thresholds, it is determined that the anchor is in a dormant period; if the anchor's off-broadcasting duration is greater than the second off-broadcasting duration threshold, it is determined that the anchor is in a loss period.

[0206] When building a model to determine the anchor's life cycle stage, two main methods are used to determine the anchor's life cycle stage. One is to determine whether a certain behavior occurs at a certain point in time, and the other is to determine whether the behavior within a certain period of time meets the conditions.

[0207] After determining the model of the anchor life cycle stages mentioned above, we constructed seven major stages of the anchor life cycle through methods such as silence recognition, stability detection, and cycle analysis. After the modeling was completed, we conducted coverage and stability tests. The results are as follows:

[0208] 1. Lifecycle label coverage 100%

[0209] Taking a certain day as an example, the composition of the anchor's life cycle stages is as follows:

[0210] Overall: Among all users who have opened live broadcast permissions, about 66% (xx million) of the anchors are old anchors who have been silent for more than 60 days (anchors in the dormant and loss periods), 7% (xx million) of the anchors are potential anchors (i.e. anchors in the introduction period), and the remaining anchors in the growth and decline periods each account for 12% (xx million), and anchors in the emerging and mature periods each account for 1.5% (xx million).

[0211] Among the active anchors in the past 4 weeks: a total of xx million anchors, 8% (xx million) are in the emerging stage, 67% (xx million) are in the growth stage, 9% (xx million) are in the mature stage, and 16% (xx million) are in the decline stage.

[0212] Weekly and daily active anchors: They are still active recently, and the probability of them being in a decline period is low. In the current version, the number of anchors in a decline period accounts for less than 0.5%.

[0213] 2. Life cycle stability is consistent with cognition

[0214] Overall, the weekly stability of the life cycle is consistent with cognition. From the perspective of several life cycles with relatively large circulation:

[0215] Emerging Stage: Mainly limited by the number of days since the first broadcast, 87% of the anchors have entered the growth stage, and 13% are still in the emerging stage (new anchors within 1 week);

[0216] Growth stage: It is very difficult for anchors in the growth stage to move up. 77% are still in the growth stage, 21% have entered the decline stage, and only 2% have moved to the mature stage.

[0217] Mature stage: Strong stability, 84% of anchors are still in the mature stage after one week, 12% have re-entered the growth stage, and 4% have entered the decline stage;

[0218] Decline stage: About 79% of the anchors are still in the decline stage, 12% of the anchors have been silent for more than 60 consecutive days and entered the dormant stage, 9% of the anchors will improve and enter the growth stage, and there are basically no anchors entering the mature stage.

[0219] Corresponding to the method for determining the anchor life cycle stage provided by the present disclosure, the present disclosure also provides a device for determining the anchor life cycle stage. Figure 10 FIG. 1 is a schematic diagram showing a structure of a device for determining a host life cycle stage according to an exemplary embodiment. Figure 10 As shown, the host life cycle stage determination device 300 includes:

[0220] The first acquisition module 302 is configured to acquire the live broadcast data of the anchor;

[0221] The classification module 304 is configured to classify the live broadcast data to obtain multiple types of live broadcast data, wherein each type of live broadcast data is used to determine a life cycle stage corresponding to the type;

[0222] The first determination module 306 is configured to determine the life cycle stage of the anchor based on multiple types of live broadcast data.

[0223] In the disclosed embodiments, since a host's live broadcast data includes multiple types of data, and different types of live broadcast data are used to determine different lifecycle stages of the host, the live broadcast data is categorized to obtain multiple types of live broadcast data. The host's lifecycle stage is then determined based on these multiple types of live broadcast data. This allows the host's lifecycle to be determined, allowing targeted measures to be taken based on the host's lifecycle.

[0224] In one or more embodiments of the present disclosure, the first determining module 306 may include:

[0225] A first determining unit is configured to determine whether the anchor is in a life cycle stage corresponding to the first type based on live broadcast data of a first type among multiple types, wherein the first type is any type among the multiple types;

[0226] The second determination unit is configured to determine whether the anchor is in the life cycle stage corresponding to the second type based on the live broadcast data of the second type among multiple types when it is determined that the anchor is not in the life cycle stage corresponding to the first type, wherein the second type is a type other than the first type among the multiple types.

[0227] In one or more embodiments of the present disclosure, the first type of live broadcast data includes the duration of the anchor's off-broadcasting;

[0228] The first determining unit is specifically configured to:

[0229] If the anchor's off-broadcasting duration is between the first off-broadcasting duration threshold and the second off-broadcasting duration threshold, it is determined that the anchor is in a dormant period;

[0230] If the anchor's off-broadcasting duration is greater than the second off-broadcasting duration threshold, it is determined that the anchor is in the churn period.

[0231] In one or more embodiments of the present disclosure, the host life cycle stage determination device 300 may further include:

[0232] The second acquisition module is configured to obtain a correspondence between the broadcaster's off-broadcast duration and the reflow rate, wherein the reflow rate represents: among M sample broadcasters who have reached the off-broadcast duration, the proportion of the number of broadcasters who resume live broadcasting within a predetermined time period to the total number of M sample broadcasters, where M is an integer greater than 1;

[0233] A third acquisition module is configured to acquire, according to the corresponding relationship, the suspension duration corresponding to the first preset reflux rate and the suspension duration corresponding to the second preset reflux rate;

[0234] The second determining module is configured to determine the suspension duration corresponding to the first preset reflux rate as a first suspension duration threshold, and to determine the suspension duration corresponding to the second preset reflux rate as a second suspension duration threshold.

[0235] In one or more embodiments of the present disclosure, the first type of live broadcast data includes a live broadcast data sequence of a host; and the first determining unit includes:

[0236] The first calculation subunit is configured to calculate a coefficient of variation of the live data sequence based on a value in the live data sequence, wherein the coefficient of variation represents a degree of dispersion of the live data sequence;

[0237] The first determining subunit is configured to determine that the anchor is in the mature stage when the coefficient of variation is within a preset range.

[0238] In one or more embodiments of the present disclosure, the host life cycle stage determination device 300 may further include:

[0239] A fourth acquisition module is configured to obtain a correspondence between a repetition rate of sample hosts and a statistical period, wherein the repetition rate represents: after N sample hosts in the mature stage are determined based on the live broadcast data sequence within the statistical period, the ratio of the number of sample hosts that continue to be in the mature stage within a preset time period to the total number of N sample hosts, where N is an integer greater than 1;

[0240] The fifth acquisition module is configured to acquire a target period corresponding to the predetermined repetition rate according to the corresponding relationship.

[0241] Among them, the classification module is specifically configured as follows:

[0242] According to the target period, the live broadcast data sequence of the anchor is extracted from the live broadcast data, wherein a value of the live broadcast data sequence is obtained based on the statistics of the live broadcast data within a target period.

[0243] In one or more embodiments of the present disclosure, the numerical values ​​in the live broadcast data sequence are the number of days the anchor has been broadcasting in different time periods; the apparatus 300 for determining the anchor's life cycle stage may further include:

[0244] a sixth acquisition module, configured to acquire data of multiple live broadcast indicators of the sample anchor;

[0245] A calculation module is configured to calculate the coefficient of variation of each live broadcast indicator based on the data of each live broadcast indicator;

[0246] A seventh acquisition module is configured to acquire a target live broadcast indicator with the smallest coefficient of variation among the multiple live broadcast indicators;

[0247] The first acquisition module is specifically configured to: acquire data of the host's target live broadcast indicator according to the target live broadcast indicator.

[0248] In one or more embodiments of the present disclosure, the first type of live broadcast data includes the host's target live broadcast time point, which is the time point when the host performs the first live broadcast, or the time point when the host performs the live broadcast during the loss period.

[0249] The first determining unit is specifically configured to:

[0250] According to the target live broadcast time point, it is determined that the anchor is in the emerging stage in the first time period after the target live broadcast time point, and it is determined that the anchor is in the growing stage in the second time period after the target live broadcast time point.

[0251] The duration of the first time period is shorter than the duration of the second time period.

[0252] In one or more embodiments of the present disclosure,

[0253] The first type of live broadcast data includes the host's first broadcast frequency, second broadcast frequency, and the host's continuous non-live broadcast duration. The first broadcast frequency refers to the host's broadcast frequency in the most recent time period, and the second broadcast frequency refers to the host's broadcast frequency in another time period before the time period.

[0254] The first determining unit is configured to:

[0255] When the first broadcast frequency is less than the second broadcast frequency, or the number of consecutive days without live broadcast is greater than a preset threshold, it is determined that the anchor is in a decline period;

[0256] When the first broadcast frequency is not less than the second broadcast frequency and the number of consecutive days without live broadcast is not greater than a preset day threshold, it is determined that the anchor is in the growth stage.

[0257] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0258] The present disclosure further provides a server, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute instructions to implement the method for determining the anchor life cycle stage in any one of the above embodiments, specifically in combination with Figure 11 Provide detailed explanation.

[0259] Figure 11 The figure is a structural block diagram of a server according to an exemplary embodiment.

[0260] like Figure 11As shown in FIG, the server can implement an exemplary hardware architecture of a server according to the information processing method and information processing apparatus in the embodiment of the present disclosure. The server can refer to the electronic device and the server in the embodiment of the present disclosure.

[0261] The server may include a processor 401 and a memory 402 storing computer program instructions. The processor 401 and the memory 402 communicate with each other via a bus 403 .

[0262] Specifically, the processor 401 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present disclosure.

[0263] Memory 402 may include a large capacity memory for information or instructions. By way of example and not limitation, memory 402 may include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 402 may include removable or non-removable (or fixed) media. Where appropriate, memory 402 may be internal or external to the integrated gateway device. In a specific embodiment, memory 402 is a non-volatile solid-state memory. In a specific embodiment, memory 402 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0264] The processor 401 reads and executes computer program instructions stored in the memory 402 to execute the method for determining the anchor life cycle stage in any of the above embodiments.

[0265] The present disclosure also provides a computer-readable storage medium including instructions, such as a memory including instructions, wherein the instructions are executable by a processor of a device to perform the above method. Alternatively, the computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, or the like.

[0266] The present disclosure also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the method for determining the anchor life cycle stage in any of the above embodiments.

[0267] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0268] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for determining the life cycle stage of a host, characterized in that: include: Get the live broadcast data of the anchor; Classifying the live broadcast data to obtain multiple types of live broadcast data, wherein each type of live broadcast data is used to determine a life cycle stage corresponding to the type, the life cycle stage including at least one of the following: dormant period, loss period, mature period, nascent period, growth period, and decline period; Determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type, wherein the first type is any type among the multiple types; If it is determined that the anchor is not in the life cycle stage corresponding to the first type, determining whether the anchor is in the life cycle stage corresponding to the second type based on live broadcast data of a second type among the multiple types, wherein the second type is a type among the multiple types other than the first type; The first type of live broadcast data includes the off-broadcast duration of the anchor; and determining whether the anchor is in a life cycle stage corresponding to the first type based on the live broadcast data of the first type among the multiple types includes: If the anchor's off-broadcasting duration is between a first off-broadcasting duration threshold and a second off-broadcasting duration threshold, it is determined that the anchor is in a dormant period; if the anchor's off-broadcasting duration is greater than the second off-broadcasting duration threshold, it is determined that the anchor is in a lost period.

2. The method according to claim 1, characterized in that The method further includes determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type: Obtaining a correspondence between the broadcaster's off-broadcast duration and the return rate, where the return rate represents: the ratio of the number of broadcasters who resume live streaming within a predetermined time period among the M sample broadcasters who have reached the off-broadcast duration to the total number of the M sample broadcasters, where M is an integer greater than 1; According to the corresponding relationship, obtaining the suspension duration corresponding to the first preset reflux rate and the suspension duration corresponding to the second preset reflux rate; The suspension duration corresponding to the first preset reflux rate is determined as the first suspension duration threshold, and the suspension duration corresponding to the second preset reflux rate is determined as the second suspension duration threshold.

3. The method according to claim 1, characterized in that The first type of live broadcast data includes a live broadcast data sequence of the anchor; The determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type includes: Calculating a coefficient of variation of the live data sequence based on values ​​in the live data sequence, wherein the coefficient of variation represents a degree of dispersion of the live data sequence; When the coefficient of variation is within a preset range, it is determined that the anchor is in the mature stage.

4. The method according to claim 3, characterized in that The method further includes determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type: Obtaining a correspondence between the repetition rate of sample hosts and the statistical period, wherein the repetition rate represents: after determining N sample hosts in the mature stage based on the live broadcast data sequence within the statistical period, the proportion of the number of sample hosts that continue to be in the mature stage within a preset time period to the total number of the N sample hosts, where N is an integer greater than 1; According to the corresponding relationship, obtaining a target period corresponding to a predetermined repetition rate; The classifying of the live broadcast data to obtain multiple types of live broadcast data includes: According to the target period, the live broadcast data sequence of the anchor is extracted from the live broadcast data, wherein a value of the live broadcast data sequence is obtained based on statistics of the live broadcast data within the target period.

5. The method according to claim 1, characterized in that Before obtaining the live broadcast data of the anchor, the method further includes: Obtain data on multiple live broadcast indicators of sample anchors; Calculate the coefficient of variation of each of the live broadcast indicators based on the data of each of the live broadcast indicators; Obtaining a target live broadcast indicator with the smallest coefficient of variation among the multiple live broadcast indicators; The method of obtaining the live broadcast data of the anchor includes: According to the target live broadcast indicator, data of the target live broadcast indicator of the anchor is obtained.

6. The method according to claim 1, wherein The first type of live broadcast data includes the host's target live broadcast time point, where the target live broadcast time point is the time point when the host performs live broadcast for the first time, or the time point when the host performs live broadcast during the churn period; The determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type includes: According to the target live broadcast time point, it is determined that the anchor is in the emerging stage in a first time period after the target live broadcast time point, and it is determined that the anchor is in the growing stage in a second time period after the target live broadcast time point. The duration of the first time period is shorter than the duration of the second time period.

7. The method according to claim 1, characterized in that The first type of live broadcast data includes the host's first broadcast frequency, second broadcast frequency, and the number of consecutive days the host has not broadcast live, wherein the first broadcast frequency is the host's broadcast frequency in a recent time period, and the second broadcast frequency is the host's broadcast frequency in another time period before the time period; The determining, based on live broadcast data of a first type among the multiple types, whether the anchor is in a life cycle stage corresponding to the first type includes: When the first broadcast frequency is less than the second broadcast frequency, or the number of consecutive days without live broadcast is greater than a preset threshold, it is determined that the anchor is in a decline period; When the first broadcast frequency is not less than the second broadcast frequency, and the number of consecutive days without live broadcast is not greater than a preset day threshold, it is determined that the anchor is in the growth stage.

8. A device for determining the life cycle stage of a host, characterized in that: include: A first acquisition module is configured to acquire live broadcast data of the anchor; a classification module configured to classify the live broadcast data to obtain a plurality of types of live broadcast data, wherein each type of live broadcast data is used to determine a life cycle stage corresponding to the type, the life cycle stage comprising at least one of the following: dormant stage, loss stage, mature stage, nascent stage, growth stage, and decline stage; A first determining module, comprising a first determining unit and a second determining unit; The first determining unit is configured to determine whether the anchor is in a life cycle stage corresponding to a first type of live broadcast data among the multiple types, wherein the first type is any type among the multiple types; The second determining unit is configured to, if it is determined that the anchor is not in the life cycle stage corresponding to the first type, determine whether the anchor is in the life cycle stage corresponding to the second type based on live broadcast data of a second type among the multiple types, wherein the second type is a type among the multiple types other than the first type; Among them, the first type of live broadcast data includes the off-broadcast duration of the anchor; the first determination unit is specifically configured to: if the off-broadcast duration of the anchor is between a first off-broadcast duration threshold and a second off-broadcast duration threshold, determine that the anchor is in a dormant period; if the off-broadcast duration of the anchor is greater than the second off-broadcast duration threshold, determine that the anchor is in a loss period.

9. The device according to claim 8, characterized in that The device further comprises: The second acquisition module is configured to obtain a correspondence between the broadcaster's off-broadcast duration and the reflow rate, wherein the reflow rate represents: the ratio of the number of broadcasters who resume live broadcasting within a predetermined time period among the M sample broadcasters who have reached the off-broadcast duration to the total number of the M sample broadcasters, where M is an integer greater than 1; A third acquisition module is configured to acquire, according to the corresponding relationship, the suspension duration corresponding to the first preset reflux rate and the suspension duration corresponding to the second preset reflux rate; The second determining module is configured to determine the suspension duration corresponding to the first preset reflux rate as the first suspension duration threshold, and determine the suspension duration corresponding to the second preset reflux rate as the second suspension duration threshold.

10. The device according to claim 8, characterized in that The first type of live broadcast data includes a live broadcast data sequence of the anchor; The first determining unit includes: a first calculation subunit configured to calculate a coefficient of variation of the live data sequence based on values ​​in the live data sequence, wherein the coefficient of variation represents a degree of dispersion of the live data sequence; The first determining subunit is configured to determine that the anchor is in the mature stage when the coefficient of variation is within a preset range.

11. The device according to claim 10, characterized in that The device further comprises: A fourth acquisition module is configured to obtain a correspondence between a repetition rate of sample anchors and a statistical period, wherein the repetition rate represents: after N sample anchors in the mature stage are determined based on the live broadcast data sequence within the statistical period, the proportion of the number of sample anchors that continue to be in the mature stage within a preset time period to the total number of the N sample anchors, where N is an integer greater than 1; A fifth acquisition module is configured to acquire a target period corresponding to the predetermined repetition rate according to the corresponding relationship; Wherein, the classification module is specifically configured as follows: According to the target period, the live broadcast data sequence of the anchor is extracted from the live broadcast data, wherein a value of the live broadcast data sequence is obtained based on statistics of the live broadcast data within the target period.

12. The device according to claim 8, characterized in that The device further comprises: a sixth acquisition module, configured to acquire data of multiple live broadcast indicators of the sample anchor; A calculation module is configured to calculate the coefficient of variation of each of the live broadcast indicators based on the data of each of the live broadcast indicators; A seventh acquisition module is configured to acquire a target live broadcast indicator with the smallest coefficient of variation among the multiple live broadcast indicators; The first acquisition module is specifically configured to: acquire data of the target live broadcast indicator of the anchor according to the target live broadcast indicator.

13. The device according to claim 8, characterized in that The first type of live broadcast data includes the host's target live broadcast time point, where the target live broadcast time point is the time point when the host performs live broadcast for the first time, or the time point when the host performs live broadcast during the churn period; The first determining unit is configured to: According to the target live broadcast time point, it is determined that the anchor is in the emerging stage in a first time period after the target live broadcast time point, and it is determined that the anchor is in the growing stage in a second time period after the target live broadcast time point. The duration of the first time period is shorter than the duration of the second time period.

14. The device according to claim 8, characterized in that The first type of live broadcast data includes the host's first broadcast frequency, second broadcast frequency, and the number of consecutive days the host has not broadcast live, wherein the first broadcast frequency is the host's broadcast frequency in a recent time period, and the second broadcast frequency is the host's broadcast frequency in another time period before the time period; The first determining unit is configured to: When the first broadcast frequency is less than the second broadcast frequency, or the number of consecutive days without live broadcast is greater than a preset threshold, it is determined that the anchor is in a decline period; When the first broadcast frequency is not less than the second broadcast frequency, and the number of consecutive days without live broadcast is not greater than a preset day threshold, it is determined that the anchor is in the growth stage.

15. A server, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the method for determining the anchor life cycle stage according to any one of claims 1 to 7.

16. A computer-readable storage medium, characterized in that When the instructions in the computer-readable storage medium are executed by the processor of the server, the server is enabled to execute the method for determining the anchor life cycle stage as described in any one of claims 1 to 7.

17. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method for determining the anchor life cycle stage according to any one of claims 1 to 7 is implemented.

Images