JP2018097393A - Engagement index value controller, engagement index value controlling method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To estimate an engagement index value even when the shape of a model for estimating an engagement index value used for the distribution control of a moving image cannot be assumed.SOLUTION: An engagement index value controller for estimating an engagement index value used for the distribution control of a moving image includes: storage means for storing an engagement model for each control parameter of the moving image; acquirement means that reads, from the storage means, the engagement model corresponding to a control parameter of a moving image as a candidate to be distributed and that acquires an engagement index value from the engagement model; and updating means that extracts, from a viewing log of a moving image, a control parameter of the moving image and an engagement index value corresponding to the control parameter and that updates, by using the engagement index value, the engagement model having been stored in the storage means and corresponding to the extracted control parameter having been extracted.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a quality control technology for communication services based on an engagement index.

  In recent years, networks have been established throughout the country, and various applications via the networks have appeared. However, at stations and event venues where people tend to gather, communication quality sufficient to use applications may not be ensured due to congestion and the like. In order to solve this problem, a method of deploying a sufficient amount of communication equipment can be considered, but since concentration of people often occurs over time, excess equipment is used in times other than the congestion time zone. This leads to inefficiency in capital investment.

  Therefore, efforts are being made to ensure quality by controlling networks and applications. At this time, what is taken as a quality index at the time of control is important. Quality indicators are broadly divided into QoS (Quality of Service) such as delay and throughput, KQI (Key Quality Indicator) such as video bit rate and number of playback stop, QoE (Quality of Experience) that the user feels, and viewing time. Can be classified into four types of engagement indexes such as the number of times of viewing and the number of times of viewing.

  Conventionally, control is often performed based on QoS and KQI indexes because of ease of measurement. However, from the viewpoint of the service providers that provide services via the network, they are more interested in quality actually felt by users, such as QoE and engagement indicators, and how they affected user behavior rather than quality on the system. It is thought that there is. Therefore, in recent years, studies on control methods based on QoE and engagement indicators are in progress.

  As a control method based on QoE, there is a technique disclosed in Non-Patent Document 1. In this technology, a control method for moving picture distribution service is proposed. First, QoE is modeled by subjective evaluation experiments. And, by maximizing the QoE in the video distribution by recommending the distribution condition that gives the highest QoE. However, when this method is applied to another service, it is necessary to perform a subjective evaluation experiment again. Therefore, it is difficult to apply this method to various services.

  On the other hand, there is a technique disclosed in Non-Patent Document 2 as a control method based on an engagement index. In this technology as well, we have proposed an engagement optimization control method for video distribution services. First, using the viewing time as an engagement index, the effect of the bit rate and playback stop time on the viewing time is modeled as a utility function. Then, the engagement index is maximized by selecting a distribution server having the highest utility function. Since this method models the engagement index based on the data, there is an advantage that it is not necessary to repeat the evaluation experiment every time the service is changed as in QoE.

Koji Yamamoto, "Introduction of efforts to realize QoE-centric operation," IEICE Technical Report, vol. 114, no. 298, Nov. 2014 X. Liu, F. Dobrian, H. Milner, J. Jiang, V. Sekar, I. Stoica, and H. Zhang. A case for a coordinated internet video control plane. In SIGCOMM, 2012.

As mentioned earlier, control based on QoE and engagement indicators is required in recent years, and quality control based on engagement indicators is particularly important in the future because service providers do not have to conduct new subjective evaluation experiments. It becomes an important technology. However, there are two problems in quality control that increases the current engagement index value.
The first problem is that the use of the same model may cause the estimation result of the engagement index value to deviate from the actual engagement index value. In Non-Patent Document 2, once a model is created, quality control is performed while continuing to use the model. However, in reality, the characteristics of the engagement index are considered to change due to changes in technology trends and service provision status. Therefore, when a certain amount of time has elapsed since the model creation, the estimation accuracy of the engagement index value may be lowered.

  The second problem is that quality control cannot be performed when the model of the engagement index cannot be assumed. As shown in Non-Patent Document 2, in order to create an engagement index estimation model, data is analyzed, the model shape (model formula) is assumed, and the coefficient is used by a method such as the least square method. In many cases, the estimation method is used. However, immediately after the service is started, the characteristics of the engagement index are not obvious, and it is difficult to assume the shape of the model. Therefore, even if control is performed, the engagement index is not necessarily maximized.

  The present invention has been made in view of the second problem described above, and estimates an engagement index value even when a model for estimating an engagement index value used for video distribution control cannot be assumed. It is an object to provide a technology that makes it possible.

According to the disclosed technology, an engagement index value control device for estimating an engagement index value used for video distribution control,
Storage means for storing an engagement model for each control parameter of the video,
An acquisition unit that reads an engagement model corresponding to a control parameter of a distribution candidate video from the storage unit, and acquires an engagement index value from the engagement model;
A video control parameter and an engagement index value corresponding to the control parameter are extracted from the video viewing log, and the extracted control parameter is stored in the storage unit using the engagement index value. An engagement index value control apparatus comprising: an updating unit that updates an engagement model to be provided.

  According to the disclosed technology, it is possible to estimate the engagement index value even when it is not possible to assume the form of a model for estimating the engagement index value used for moving image distribution control.

It is a system configuration figure in an embodiment of the invention. It is a hardware block diagram of the apparatus which comprises the system in embodiment of this invention. 4 is a flowchart illustrating a procedure for updating an engagement model in the first embodiment. 3 is a flowchart illustrating a procedure for estimating an engagement index value in the first embodiment. It is a flowchart which shows the procedure of the engagement model update in Example 2. FIG. It is a figure which shows the example of the table stored in engagement model DB202 in Example 2. FIG. It is a flowchart which shows the procedure of estimation of the engagement index value in Example 2.

  Hereinafter, an embodiment (this embodiment) of the present invention will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment.

(Overview)
In the present embodiment, a technique for solving the first problem described above is described as a first example, and a technique for solving the second problem is described as a second example. However, it is not essential to solve both the first problem and the second problem by the technique according to the present embodiment. For example, one problem may be solved by the technique according to the present embodiment, and the other problem may be solved by another technique, or only one problem may be solved.

  With regard to the engagement index that can assume the model form (model formula), the first problem is that the estimation result of the engagement index value may deviate from the actual engagement index value by continuing to use the same model. On the other hand, in the first embodiment, the problem is solved by sequentially updating the model formula (estimation formula) of the engagement index. First, the service provider determines a parameter for controlling the quality of the service and an engagement index to be estimated according to the service to be provided. Next, data analysis is performed, and an engagement index estimation model including a coefficient is created. However, the coefficient needs to be a number obtained from an expression represented by data statistics. Then, the initial coefficient of the engagement index estimation model is determined from already accumulated data using the least square method or the like. While providing services based on the model constructed in this way, feedback information is collected from the user, and statistics included in the coefficients are updated. Then, the engagement index estimation model is periodically updated based on the updated statistics. In this method, since the model can be updated as needed based on feedback information from the user, a model adapted to the technical trend and the service provision status can be constructed. By performing quality control based on the estimation result obtained in this way, it is possible to realize more accurate optimization for the engagement index that can assume the shape of the model.

  The second problem that quality control cannot be performed when the shape of the engagement index model cannot be assumed is solved in the second embodiment by performing non-parametric modeling. First, the service provider determines a parameter for controlling the quality of the service and an engagement index to be estimated according to the service to be provided. Then, a histogram is created according to the combination of each parameter, and the representative value is adopted as the estimated value. When estimating an engagement index value related to a certain control parameter, a histogram parameter is acquired by inputting the control parameter, and a desired value is acquired by calculating a representative value therefrom. In this method, since it is not necessary to assume the shape of the model, the model can be constructed without analyzing the data as in the conventional method. By performing quality control based on the estimation result obtained in this way, it is possible to realize the optimization of an engagement index that cannot assume the model shape.

(System configuration, overall operation)
FIG. 1 shows a system configuration in the present embodiment (common to Examples 1 and 2). As shown in the figure, the system according to the present embodiment includes an engagement estimation unit 100, an engagement model update unit 200, a bit rate selection unit 300, a distribution server 400, and a terminal 500.

  The engagement estimation unit 100 is a functional unit that estimates (calculates) an engagement index value, and includes a reproduction stop rate estimation unit 101 and an engagement calculation unit 102. The engagement model update unit 200 is a functional unit that updates the engagement model, and includes a model update unit 201 and an engagement model DB 202.

  The terminal 100 has a video playback application. In the present embodiment, it is assumed that terminal 500 is a mobile terminal that communicates by connecting to a mobile network. However, the present invention is not limited to this, and a fixed terminal that communicates by connecting to a fixed network. It may be. In addition, although one terminal 500 is illustrated in FIG. 1, there are actually a large number of terminals, and the processing described below is performed on each of them.

  The processing contents of each unit and each unit will be described in the subsequent processing description. In the present embodiment, a description will be given by taking a video distribution service as a target communication service and viewing time as a corresponding engagement index. However, the application destination of the present invention is not limited to this, and as an engagement index, for example, any one or more of the usage time of the entire service, the rate of transition to the billing plan, the service cancellation rate, etc. It may be adopted.

  First, with reference to FIG. 1, an overall processing flow of motion picture distribution control (quality control) in the present embodiment (common to Examples 1 and 2) will be described.

  Step S1) The terminal (video playback application) 500 makes a video distribution request to the distribution server 400. Communication environment information (for example, a set of time zone, carrier, and cell ID) is given to the distribution request. However, this communication environment information is an example, and in the case of a fixed network, the ISP or the like used may be included in the elements of the communication environment information.

  Step S2) The distribution server 400 notifies the bit rate selection unit 300 of the distributable bit rate sequence of the moving image and the communication environment information of the user (terminal 500). The distributable bit rate sequence is a list of bit rates that can be distributed by the distribution server 400, such as 100 Kpbs, 200 Kbps, 500 Kbps,.

  Step S3) The bit rate selection unit 300 requests the engagement estimation unit 100 for an estimated engagement index value for a certain combination of communication environment and bit rate by transmitting the communication environment information and the bit rate. The method for estimating the engagement index value will be described in detail in each embodiment. Step S3 is performed for each bit rate in the distributable bit rate sequence.

  Step S4) The bit rate selection unit 300 acquires an engagement index value (estimated value) when distributed at each distributable bit rate from the engagement estimation unit 100.

  Step S5) The bit rate selection unit 300 notifies the distribution server 400 of the bit rate with the highest engagement index value as the optimum distribution bit rate.

  Step S6) The distribution server 400 distributes the moving image using the acquired optimum distribution bit rate.

  Step S7) After the end of moving image viewing, the terminal 500 sends a viewing log (eg, a set of bit rate, playback stop rate, viewing time) to the distribution server 400. Note that the terminal 500 may send the viewing log to the engagement model update unit 200.

  Step S8) The distribution server 400 sends the viewing log to the engagement model update unit 200. Then, the engagement model update unit 200 updates the engagement model based on the acquired viewing log. Note that the engagement model may be updated every time a viewing log is acquired, periodically using a timer, or at a timing other than these. The method for updating the engagement model will be described in detail in each embodiment.

(Hardware configuration)
Each of the engagement estimation unit 100, the engagement model update unit 200, and the bit rate selection unit 300 shown in FIG. 1 may be a single device, or the engagement estimation unit 100, the engagement model update unit 200, and the bit rate. The selection unit 300 may constitute one device, or the engagement estimation unit 100 and the engagement model update unit 200 may constitute one device, or a device configuration other than these.

  As an example, a configuration including the engagement estimation unit 100 and the engagement model update unit 200 can be referred to as an engagement index value control device. The engagement index estimation device can be configured by one or a plurality of computers, for example.

  Each device such as the engagement index value control device described above can be realized by causing one or a plurality of computers to execute a program describing the processing contents described in the present embodiment. In other words, each function of the device can be realized by executing a program corresponding to processing executed in the device using hardware resources such as a CPU and a memory built in the computer. . The above-mentioned program can be recorded on a computer-readable recording medium (portable memory or the like), stored, or distributed. It is also possible to provide the program through a network such as the Internet or electronic mail.

  FIG. 2 is a diagram illustrating a hardware configuration example when the apparatus is realized by a computer. 2 includes a drive device 150, an auxiliary storage device 152, a memory device 153, a CPU 154, an interface device 155, a display device 156, an input device 157, and the like that are mutually connected by a bus B.

  A program for realizing processing in the apparatus is provided by a recording medium 151 such as a CD-ROM or a memory card, for example. When the recording medium 151 storing the program is set in the drive device 150, the program is installed from the recording medium 151 into the auxiliary storage device 152 via the drive device 150. However, the program does not necessarily have to be installed from the recording medium 151, and may be downloaded from another computer via a network. The auxiliary storage device 152 stores the installed program and also stores necessary files and data.

  The memory device 153 reads the program from the auxiliary storage device 152 and stores it when there is an instruction to start the program. The CPU 154 realizes functions related to the device in accordance with a program stored in the memory device 153. The interface device 155 is used as an interface for connecting to a network. The display device 156 displays a GUI (Graphical User Interface) or the like by a program. The input device 157 includes a keyboard and mouse, buttons, a touch panel, and the like, and is used to input various operation instructions.

  Hereinafter, updating of the engagement model and estimation of the engagement index value will be described in detail. A case where the shape of the engagement index model can be assumed will be described as a first embodiment, and a case where the shape of the engagement index model cannot be assumed will be described as a second embodiment. The system illustrated in FIG. 1 may have a function of executing both the first and second embodiments, or may have a function of any one of the first and second embodiments. . When a function for executing both the first and second embodiments is provided, for example, according to the type of service, the control according to the first embodiment and the control according to the second embodiment may be properly used. Depending on the control, the control of the first embodiment and the control of the second embodiment may be properly used.

(Example 1: When the shape of the engagement index model can be assumed)
<Updating the engagement model of Example 1>
In this embodiment, the following utility function shown in Non-Patent Document 2 is used as an engagement model expression.

Utility = c × BuffRatio + d × bitrate
Here, BuffRatio represents the ratio of the stop time to the playback time (video length + stop time), and bitrate represents the delivery bit rate. BuffRatio and bitrate may be collectively referred to as control parameters. In Non-Patent Document 2, c = −3.7 and d = 1/20, but in Example 1, this coefficient is updated. The engagement model is updated by the engagement model updating unit 200 according to the procedure shown in FIG.

  Step S101) First, the model update unit 201 of the engagement model update unit 200 acquires the viewing log transmitted from the distribution server 400. The viewing log may be received from the user terminal 500.

  Step S102) The model update unit 201 extracts a bit rate, a playback stop ratio (BuffRatio), and a viewing time (viewtime) at that time from the viewing log.

Step S103) Then, using the extracted values, based on the following equations, the number c ₁ , the number c ₂ , the number c ₃ , the number d ₁ , the number d ₂ , the number d ₃ , the number e, the number n, and the number c The coefficients c and d are obtained by calculating the number d sequentially from the top.

c ₁ ← c ₁ + BuffRatio
c ₂ ← c ₂ + BuffRatio ²
c ₃ ← c ₃ + BuffRatio × viewtime
d ₁ ← d ₁ + bitrate
d ₂ ← d ₂ + bitrate ²
d ₃ ← d ₃ + bitrate × viewtime
e ← e + viewtime
n ← n + 1
c ← (n × c ₃ − c ₁ × e) / (n × c ₂ − c ₂ ² )
d ← (n × d ₃ − d ₁ × e) / (n × d ₂ − d ₂ ² )
Each calculated number is stored in the engagement model DB 202. The above calculation corresponds to the calculation for obtaining c and d in the Utility model equation by the least square method using the measurement values (bitrate, BuffRatio, etc.) obtained sequentially. That is, the coefficient is updated so that the Utility model formula is fitted to the value extracted from the viewing log.

More specifically, the number c ₁ , the number c ₂ , the number c ₃ , the number d ₁ , the number d ₂ , the number d ₃ , the number e, the number n, the number c, the number Initial coefficients c and d are obtained by performing the above calculation with the initial value of each number of d being 0. Then, while providing a service using the model to which the initial coefficient is applied, feedback information (viewing log) from the user is collected, and c and d are updated by performing the above calculation.

<Estimation of Engagement Index Value of Example 1>
In the first embodiment, the engagement estimation unit 100 performs the estimation (calculation) of the engagement index value according to the procedure shown in FIG.

Step S111) First, the reproduction stop ratio estimation unit 101 estimates the reproduction stop ratio (BuffRatio) for viewing based on the communication environment information and the bit rate received in step S3 of FIG. BuffRatio estimation itself can be performed with existing technology. For example, using the technique of Non-Patent Document 1, first, the throughput is estimated from the communication environment information, and the reproduction stop time T _stop is estimated using the estimated throughput and the bit rate as inputs. Then, the video length is set as T _length, and BuffRatio is calculated by the following formula.

BuffRatio = T _stop / (T _length + T _stop )
Step S112) Next, the engagement calculation unit 102 calculates an engagement index value. First, the engagement model DB 202 is accessed to obtain the numbers c and d. The access to the engagement model DB 202 and the acquisition of the numbers c and d correspond to “model request” and “model response” between the engagement calculation unit 102 and the engagement model DB 202 in FIG.

  Then, the engagement calculation unit 102 calculates a desired engagement index value by inputting the BuffRatio estimated in step S111 and the corresponding bit rate into the utility function (model formula).

(Example 2: When the shape of the engagement index model cannot be assumed)
Next, a description will be given of a second embodiment which is an embodiment in the case where the shape of the engagement index model cannot be assumed.

<Updating the engagement model of Example 2>
The engagement model update in this embodiment is performed by the engagement model update unit 200 in the procedure shown in FIG.

  In the present embodiment, in the engagement model DB 202, past engagement index value data is accumulated in the form of, for example, a table for each control parameter (specifically, BuffRatio and bitrate). Then, the model update unit 201 of the engagement model update unit 200 sequentially updates the data of the engagement index value based on the viewing log.

  In step S201 of FIG. 5, the model update unit 201 receives the viewing log. In step S202, the count of the corresponding engagement index value in the table stored in the engagement model DB 202 is updated. Specifically, it is as follows.

  FIG. 6 shows an example of a table stored in the engagement model DB 202. In this example, the left two columns represent the control parameters and the remaining eight columns represent the count values of the past engagement index values, and the past engagement index values in the state of each control parameter according to the range indicated in the first row. Has been counted. This range may be freely determined by each business operator according to the nature of the service. In addition, instead of setting the range as shown in FIG. 6 and counting, the representative value of the engagement index value corresponding to each control parameter (eg, average value, weighted average value, percentage value, median value, maximum value) Only frequent values etc.) may be calculated and stored, and the values may be updated as needed. This representative value is also an example of an engagement model.

  Each of the second and subsequent lines in FIG. 6 is referred to as an engagement model in the control parameter in the present embodiment, and when the engagement model is updated, the corresponding line is updated. For example, if the viewing time for watching a video with bitrate = 200kbps and BuffRatio = 0.24 was 50 seconds, and information that the throughput was 901kbps was recorded, the 8th row and the 5th column (bitrate = 200kbps, BuffRatio = 0.2 to 0.3 The count value in the range (columns 41 to 60 in the row) is updated from 30 to 31. This update process may be performed every time the viewing log data arrives, or the viewing log data may be temporarily stored and the updating process may be performed at a specified time.

<Estimation of Engagement Index Value of Example 2>
In the second embodiment, the engagement estimation unit 100 performs the estimation (calculation) of the engagement index value according to the procedure shown in FIG.

  Step S211 is the same as step S111 in the first embodiment. That is, the reproduction stop ratio estimation unit 101 estimates the reproduction stop ratio (BuffRatio).

  In step S212, the engagement calculation unit 102 accesses the engagement model DB 202 and acquires an engagement model corresponding to the target BuffRatio and bit rate (model request, model response in FIG. 1). Then, the engagement calculation unit 102 obtains representative values (eg, average value, weighted average value, percentage value, median value, mode value) of the engagement index value from the acquired engagement model (the column of the count value of the engagement index value). Etc.), and this representative value is used as an engagement index value corresponding to the corresponding bit rate.

(Summary of embodiment)
As described above, according to the present embodiment, an engagement index value control apparatus for estimating an engagement index value used for moving image distribution control, and a predetermined model formula for calculating the engagement index value A storage unit for storing a coefficient applied to the storage unit, and reading the coefficient from the storage unit, and inputting the control parameter of the moving image of the distribution candidate into the predetermined model formula to which the coefficient is applied, thereby determining the engagement index value. A calculation unit for calculating, and a control unit for extracting the control parameter of the video from the viewing log of the video, and an update unit for updating the coefficient stored in the storage unit using the control parameter. An engagement index value control device is provided.

  The engagement model DB 202, the engagement calculation unit 102, and the model update unit 201 described in the first embodiment are examples of a storage unit, a calculation unit, and an update unit, respectively.

  For example, the updating unit updates the coefficient so that the predetermined model formula is fitted to a value extracted from the viewing log. The control parameter is, for example, a bit rate and a reproduction stop ratio, and the engagement index value is, for example, viewing time.

  In addition, according to the present embodiment, an engagement index value control apparatus for estimating an engagement index value used for moving image distribution control, the storage unit storing an engagement model for each control parameter of the moving image, and distribution An engagement model corresponding to the control parameter of the candidate video is read from the storage unit, an acquisition unit that acquires an engagement index value from the engagement model, a video control parameter from the video viewing log, and the control parameter An engagement index value, and an update means for updating an engagement model stored in the storage means and corresponding to the extracted control parameter using the engagement index value. Value control device It is provided.

  The engagement model DB 202, the engagement calculation unit 102, and the model update unit 201 described in the second embodiment are examples of a storage unit, an acquisition unit, and an update unit, respectively.

  The engagement model is, for example, a histogram composed of count values for each range of engagement index values, and the updating unit updates the count values corresponding to the extracted engagement index values. The control parameter is, for example, a bit rate and a reproduction stop ratio, and the engagement index value is, for example, viewing time.

(Effect of embodiment)
According to the embodiment of the present invention, engagement index optimization control can be performed by performing non-parametric modeling for a service that cannot assume the form of an engagement index estimation model, and the model form is assumed. By sequentially updating models for possible services, it is possible to perform highly accurate engagement index optimization control that follows changes in technology trends and service provision status.

  Although the present embodiment has been described above, the present invention is not limited to the specific embodiment, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims. Is possible.

DESCRIPTION OF SYMBOLS 100 Engagement estimation part 101 Reproduction | regeneration stop ratio estimation unit 102 Engagement calculation unit 200 Engagement model update part 201 Model update unit 202 Engagement model DB
300 Bit rate selection unit 400 Distribution server 500 Terminal 150 Drive device 151 Recording medium 152 Auxiliary storage device 153 Memory device 154 CPU
155 Interface device 156 Display device 157 Input device

Claims (7)

An engagement index value control device for estimating an engagement index value used for video distribution control,
Storage means for storing an engagement model for each control parameter of the video,
An acquisition unit that reads an engagement model corresponding to a control parameter of a distribution candidate video from the storage unit, and acquires an engagement index value from the engagement model;
A video control parameter and an engagement index value corresponding to the control parameter are extracted from the video viewing log, and the extracted control parameter is stored in the storage unit using the engagement index value. An engagement index value control device comprising: an updating unit that updates an engagement model to be updated. The said engagement model is a histogram which consists of a count value for every range of an engagement index value, and the said update means updates the count value corresponding to the said extracted engagement index value. Engagement index value control device. The engagement index value control apparatus according to claim 1, wherein the control parameters are a bit rate and a playback stop ratio, and the engagement index value is a viewing time. An engagement index value control method executed by an engagement index value control device for estimating an engagement index value used for video distribution control,
The engagement index value control device includes storage means for storing an engagement model for each control parameter of a moving image,
An acquisition step of reading an engagement model corresponding to the control parameter of the distribution candidate video from the storage unit, and acquiring an engagement index value from the engagement model;
A video control parameter and an engagement index value corresponding to the control parameter are extracted from the video viewing log, and the extracted control parameter is stored in the storage unit using the engagement index value. An engagement index value control method comprising: an updating step for updating an engagement model to be performed. The engagement model is a histogram composed of count values for each range of engagement index values, and in the updating step, the engagement index value control device updates a count value corresponding to the extracted engagement index values. 5. The engagement index value control method according to claim 4, wherein the engagement index value is controlled. The engagement index value control method according to claim 4 or 5, wherein the control parameters are a bit rate and a reproduction stop ratio, and the engagement index value is a viewing time.   The program for functioning a computer as each means in the engagement index value control apparatus of any one of Claims 1 thru | or 3.

Images