JP2017194756A - Web browsing quality estimation device, Web browsing quality estimation method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately estimate a total Quality of Experience (QoE) that is a total quality after multiple pages are shifted in Web browsing.SOLUTION: A Web browsing quality estimation device comprises: an acquisition unit for acquiring QoE estimates, for each of multiple Web pages of a shift target, based on a display waiting time from generation of a display instruction of the Web page to display of the Web page; a first calculation unit for calculating an average value of the multiple acquired QoE estimates; a second calculation unit for calculating variations in the multiple QoE estimates acquired by the acquisition unit; and a third calculation unit for calculating an estimate of a Web browsing QoE that is a QoE after the multiple Web pages are shifted on the basis of the average value and the variations.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a web browsing quality estimation apparatus, a web browsing quality estimation method, and a program.

  The use of Web applications that are used via networks such as the Internet and intranets, and applications that use cloud networks is progressing.

  In order to provide a Web browsing service that satisfies the user, it is necessary to appropriately manage QoE (Quality of Experience) for Web browsing. As a factor of QoE reduction for Web browsing, there is a Web page display waiting time. Here, the display waiting time is defined as “the time from when the user makes a processing request regarding the Web browsing service at the Web browsing receiving terminal until the processing result is displayed on the screen of the Web browsing receiving terminal”. However, the processing request includes a request automatically performed by the user's terminal in addition to a request intentionally performed by the user.

  In Non-Patent Document 1, as a method for estimating the QoE with respect to the display waiting time, the time until the server response returns from the user processing request and the processing result are displayed on the user terminal after the server response returns to the user terminal. A method for estimating the QoE based on the time until is disclosed.

International Telecommunication Union, "Estimating end-to-end performance in ip networks for data applications", ITU-T Recommendation G.1030, November 2005. Tobias Hosfeld, Raimund Schatz, Sebastian Biedermann, Alexander Platzer, Sebastian Egger, and Markus Fiedler, "The Memory Effect and Its Implications on Web QoE Modeling," In 23rd International Teletraffic Congress (ITC 23), San Francisco, USA, September 2011.

  In Web browsing, for example, when a user views news, when the user requests (starts) a top page of a news site, the top page is displayed. Next, when the user clicks the first article title, the article text is displayed. Furthermore, when the user clicks on a category of an article, a list of articles in that category is displayed. Next, when the user clicks the article title, the second article text is displayed.

  In this way, the user achieves the purpose by changing a plurality of Web pages. Therefore, in order to estimate the Web browsing QoE, it is necessary to grasp not only the QoE for each Web page but also the comprehensive QoE until the user achieves the purpose and finishes the Web browsing.

  As a method for estimating the total QoE after transitioning a plurality of Web pages, a model that is estimated using an average value of QoE for each Web page can be considered, but perceptual characteristics with respect to QoE fluctuation cannot be considered.

  In Non-Patent Document 2, the storage model is applied to the QoE fluctuation for each Web page, but the influence of the QoE fluctuation on the overall QoE is not considered.

  The present invention has been made in view of the above points, and an object of the present invention is to accurately estimate the overall QoE, which is the overall quality after a plurality of pages are transitioned by Web browsing.

  Therefore, in order to solve the above problem, the Web browsing quality estimation apparatus has a display waiting time from when an instruction to display the Web page is generated for each of a plurality of Web pages to be transitioned until the Web page is displayed. An acquisition unit that acquires a QoE (Quality of Experience) estimation value based thereon, a first calculation unit that calculates an average value of a plurality of QoE estimation values acquired by the acquisition unit, and a plurality of acquisition units acquired by the acquisition unit A second calculating unit that calculates a variation of the estimated QoE value, and a third calculating unit that calculates an estimated value of Web browsing QoE that is a QoE after transitioning the plurality of Web pages based on the average value and the variation. And a calculation unit.

  It is possible to accurately estimate the overall QoE, which is the overall quality after transitioning a plurality of pages by Web browsing.

It is a figure which shows the system configuration example in embodiment of this invention. It is a figure which shows the hardware structural example of the web browsing receiving terminal in embodiment of this invention. It is a figure which shows the function structural example of the web browsing receiving terminal in embodiment of this invention. It is a flowchart for demonstrating an example of the process sequence which a quality estimation part performs. It is a figure which shows the example of data for demonstrating the process sequence which a quality estimation part performs. It is a figure which shows the estimation precision regarding Web browsing quality estimation in embodiment of this invention.

  Hereinafter, embodiments 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.

(System configuration, device configuration)
FIG. 1 is a diagram showing an example of a system configuration in the embodiment of the present invention. In FIG. 1, one or more Web servers 20 and one or more Web browsing receiving terminals 10 are connected via an IP network 30 such as a LAN (Local Area Network) or the Internet.

  The web server 20 is a server that provides a web browsing service. A web browsing service refers to a service provided via a web page. That is, the Web server 20 transmits the requested Web page to the request source. Although only one Web server 20 is shown in FIG. 1 for the sake of simplification of explanation, more Web servers 20 than these may be connected to the IP network 30.

  The Web browsing receiving terminal 10 is, for example, a PC (Personal Computer), a PDA (Personal digital Assistant), a feature phone, a smartphone (Smartphone), or a tablet terminal. The Web browsing receiving terminal 10 only needs to have a Web browsing service processing function, and the mode is not limited to a specific one. In FIG. 1, only one Web browsing receiving terminal 10 is shown for simplification of explanation, but more Web browsing receiving terminals 10 may be connected to the IP network 30.

  FIG. 2 is a diagram illustrating a hardware configuration example of the Web browsing receiving terminal according to the embodiment of the present invention. 2 includes a drive device 100, an auxiliary storage device 102, a memory device 103, a CPU 104, an interface device 105, a display device 106, an input device 107, and the like, which are mutually connected by a bus B. .

  A program for realizing processing in the Web browsing receiving terminal 10 is provided by a recording medium 101 such as a CD-ROM or a memory card. When the recording medium 101 storing the program is set in the drive device 100, the program is installed from the recording medium 101 to the auxiliary storage device 102 via the drive device 100. However, the program need not be installed from the recording medium 101 and may be downloaded from another computer via a network. The auxiliary storage device 102 stores the installed program and also stores necessary files and data.

  The memory device 103 reads the program from the auxiliary storage device 102 and stores it when there is an instruction to start the program. The CPU 104 realizes functions related to the web browsing receiving terminal 10 according to a program stored in the memory device 103. The interface device 105 is used as an interface for connecting to a network. The display device 106 displays a GUI (Graphical User Interface) or the like by a program. The input device 107 includes a keyboard and mouse, buttons, a touch panel, and the like, and is used to input various operation instructions.

  FIG. 3 is a diagram illustrating a functional configuration example of the Web browsing receiving terminal according to the embodiment of the present invention. In FIG. 3, the Web browsing receiving terminal 10 includes a display waiting time acquisition unit 11 and a quality estimation unit 12. The quality estimation unit 12 estimates a quality of experience (QoE (Quality of Service)) after transition of a plurality of Web pages for a Web browsing service realized via the IP network 30.

  In FIG. 3, the quality estimation unit 12 includes a QoE acquisition unit 121, a QoE average value calculation unit 122, a QoE variation calculation unit 123, a QoE estimation model storage unit 124, a web browsing QoE estimation value calculation unit 125, and the like. Each of these units is realized by processing that one or more programs installed in the Web browsing receiving terminal 10 cause the CPU 104 to execute. Note that the quality estimation unit 12 is realized by a program and the CPU 104 as an example. For example, the quality estimation unit 12 may be realized as hardware by a dedicated logic circuit.

  The display waiting time acquisition unit 11 acquires the display waiting time of each of a plurality of Web pages to be transitioned. In the present embodiment, the display waiting time is “the time from when the user makes a processing request regarding the Web browsing service at the Web browsing receiving terminal 10 until the processing result is displayed on the screen of the Web browsing receiving terminal 10”. It is. However, the processing request may be automatically generated by the Web browsing receiving terminal 10 other than the one intentionally performed by the user. Therefore, the display wait time is an elapsed time from when a web page display instruction is generated until the web page is displayed. Note that the display waiting time may be defined as “time from when the user executes the processing of the Web browsing service until the processing response is displayed on the terminal” in the same sense.

  The QoE acquisition unit 121 calculates an estimated value of QoE for each Web page based on the display waiting time of each Web page acquired by the display waiting time acquisition unit 11. Note that the QoE acquisition unit 121 may be changed to using the estimated value of QoE for each Web page generated by other means.

  The QoE average value calculation unit 122 calculates an average value for the estimated value of QoE for each Web page.

  The QoE variation calculation unit 123 calculates QoE variation. More specifically, the QoE variation calculation unit 123 calculates, for example, the difference between the previous and next QoEs, and calculates the average as the variation.

  The QoE estimation model storage unit 124 stores a model formula for estimating the Web browsing QoE using the QoE average value and the variation of the QoE. Specifically, the QoE estimation model storage unit 124 stores a QoE estimation formula and a coefficient applied to the QoE estimation formula. The contents of the formula will be described later.

  The Web browsing QoE estimated value calculation unit 125 includes a QoE average value calculated by the QoE average value calculation unit 122, a QoE variation (such as an average value of QoE differences) calculated by the QoE variation calculation unit 123, and a QoE estimation model. An estimated value of Web browsing QoE (hereinafter referred to as Web browsing QoE) is calculated using the QoE estimation formula and coefficient stored in the storage unit 124.

  Web browsing QoE refers to QoE after a plurality of Web pages having different display wait times from each other. Based on the subjective quality evaluation method for Web browsing as described in Japanese Patent Application No. 1501. It is obtained in a subjective quality evaluation experiment as a mean opinion score (MOS) with respect to a score obtained by a five-level quality scale (ACR) as described in 910.

(Processing procedure)
Hereinafter, a processing procedure executed by the quality estimation unit 12 will be described. FIG. 4 is a flowchart for explaining an example of a processing procedure executed by the quality estimation unit 12. FIG. 4 is a flowchart on the assumption that the display waiting time of the Web page has already been acquired for each displayed (transitioned) Web page. FIG. 5 is a diagram illustrating an example of data for explaining a processing procedure executed by the quality estimation unit.

As the premise process, the display wait time acquisition unit 11 displays the display wait time of the web page for various processes when the user executes the web browsing using the web browsing receiving terminal 10 (transitioned). ) Acquire for each Web page. In FIG. 5, the Web pages to be displayed are changed in the order of the top page, news 1, category, and news 2, and the display waiting time (wt _i ) of each Web page is 3 seconds, 2 seconds, 1 second, 2. An example of 5 seconds is shown. In FIG. 5, i means the i-th Web page in the display order. N indicates the number of Web pages.

  The display waiting time of the Web page may be calculated using a log such as an HTTP request in the Web server 20, for example. In this case, the display wait time acquisition unit 11 acquires the calculation result by the Web server 20 from the Web server 20. Alternatively, the display wait time may be acquired using an existing technology such as the navigation timing API of Javascript (registered trademark).

  In the flowchart of FIG. 4, the QoE acquisition unit 121 first acquires an estimated value of QoE for each Web page based on the display waiting time acquired for the Web page (Step S <b> 101). The estimation of QoE for each Web page based on the display waiting time is, for example, ITU-T recommendation G.E. It may be performed using an existing technique such as an estimation formula described in Appendix 30 of 1030. Alternatively, a plurality of Web pages having different display wait times are created, and QoE for the display wait time of the Web page is calculated by, for example, ITU-T Recommendation The relationship between the Web page display latency and the average opinion score (DMOS) obtained by DCR, which is obtained by a rating based on a five-stage disturbance category rating (DCR) as described in 910 May be formulated as a new QoE estimation model.

In FIG. 5, each estimated QoE value DMOS _i for each Web page acquired in step S101 is indicated by a black circle (●).

Subsequently, the QoE average value calculation unit 122 calculates a QoE average value for the obtained estimated values DMOS _i of the plurality of QoEs (hereinafter referred to as “QoE (DMOS _i )”), and calculates the QoE variation calculation unit 123. Finds variations in QoE (DMOS _i ) for each page (step S102). The variation in the present embodiment is an absolute value average with respect to the difference between the previous and subsequent QoE values, but may be obtained as an absolute value average of a decrease value only when QoE is lowered.

In FIG. 5, the absolute value of the difference between QoE (DMOS _{i + 1} ) and QoE (DMOS _i ) is shown as D _{i + 1} , and the absolute value of the difference between QoE (DMOS _{i + 2} ) and QoE (DMOS _{i + 1} ) is shown as D _{i + 2.} , The absolute value of the difference between QoE (DMOS _n ) and QoE (DMOS _{i + 2} ) is shown as D _n . When the absolute value average with respect to the difference between the QoE values before and after is regarded as variation, the variation is calculated by taking these averages. As an example, QoE (DMOS _i ) decreases from QoE (DMOS _i ) by A to QoE (DMOS _{i + 1} ), the next QoE (DMOS _{i + 2} ) increases from QoE (DMOS _{i + 1} ), and the next QoE (DMOS _n ) Is smaller than QoE (DMOS _{i + 2} ) by B, and if the absolute value average of the decreased values only when QoE decreases is varied, the average of A and B varies.

  As described above, the QoE estimation model storage unit 124 stores a QoE estimation formula and a coefficient as a model for estimating the Web browsing QoE using the QoE average value and the variation of the QoE. Note that the QoE estimation formula in the present embodiment is an example of a formula that represents the relationship between the QoE average value, QoE variation, and Web browsing QoE.

  The Web browsing QoE estimated value calculation unit 125 calculates Web browsing QoE when browsing a plurality of Web pages to be finally obtained (step S103).

  An example of the QoE estimation formula for calculating the Web browsing QoE (MOS) using the QoE average value (avgDMOS) and the variation (avgD) of QoE is shown below as Formula (1).

MOS = a ₀ + a ₁ avgDMOS + a ₂ avgD Equation (1)
The coefficients a ₀ , a ₁ and a ₂ in the above equation (1) are constants. In step S103, the web browsing QoE estimated value calculation unit 125 applies the QoE average value (avgDMOS) calculated in step S102 and the QoE variation (avgD) to the coefficients a ₀ , a ₁ , and a ₂ that are the above constants. The web browsing QoE is calculated by inputting the formula (1). For example, the Web browsing QoE (MOS) is calculated as a value obtained by correcting the QoE average value (avgDMOS) to be smaller as the QoE variation (avgD) is larger.

  Formula (1) is an example, and formulas other than Formula (1) are used as QoE estimation formulas for calculating Web browsing QoE (MOS) using the QoE average value (avgDMOS) and QoE variation (avgD). May be used. In the present embodiment, the Web browsing QoE estimated value calculation unit 125 reads out and uses the QoE estimation formula and coefficient from the QoE estimation model storage unit 124, but this is an example. For example, the Web browsing QoE estimated value calculation unit 125 may include processing logic corresponding to a QoE estimation formula and a coefficient without including the QoE estimation model storage unit 124.

FIG. 6 is a diagram showing the estimation accuracy related to Web browsing quality estimation in the embodiment of the present invention. FIG. 6 shows the estimated MOS (that is, estimated Web browsing QoE) obtained by the equation (1) for the actual measurement MOS (Web browsing QoE). In the figure, Pearson's product-moment correlation coefficient (R) and estimated accuracy RMSE (Root Mean Square Error) are shown in the figure. The coefficients a ₀ , a ₁ , and a ₂ in Equation (1) can be optimized so that the correlation coefficient R is as high as possible and the RMSE is as small as possible. In FIG. 6, the closer the plot is to the 45 ° oblique line, the higher the accuracy of the estimation MOS.

(Effect of embodiment)
As described above, according to the present embodiment, Web browsing QoE is estimated in consideration of psychological characteristics related to variations in QoE. As a result, it is possible to estimate the total QoE with higher accuracy than in the prior art. Moreover, the quality of the service provided to the user can be maintained and improved by improving the quality or equipment using the estimation result of Web browsing QoE.

(Other examples)
In the above description, an example in which the Web browsing receiving terminal 10 includes the quality estimation unit 12 has been described. However, a device other than the Web browsing receiving terminal 10 and connected to the IP network 30 includes the quality estimation unit 12. May be. In this case, the quality estimation unit 12 may acquire display wait time information when the user performs Web browsing from the Web browsing receiving terminal 10 via the network. In such a configuration, QoE can be grasped as the usage status of Web traffic used in the network.

  In addition, the Web server 20 may have the quality estimation unit 12. In this case, the QoE for the Web service provided by the Web server 20 can be grasped.

  Therefore, the flowchart shown in FIG. 4 may be executed by a CPU of a device other than the Web browsing receiving terminal 10.

  In the present embodiment, the Web browsing receiving terminal 10 (quality estimation unit 12) is an example of a Web browsing quality estimation apparatus. The QoE acquisition unit 121 is an example of an acquisition unit. The QoE average value calculation unit 122 for each page is an example of a first calculation unit. The QoE variation calculation unit 123 is an example of a second calculation unit. The web browsing QoE estimated value calculation unit 125 is an example of a third calculation unit.

  As mentioned above, although the Example of this invention was explained in full detail, this invention is not limited to the said embodiment. The above embodiment can be variously modified and implemented without departing from the gist of the present invention. In short, the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. Further, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

DESCRIPTION OF SYMBOLS 10 Web browsing receiving terminal 11 Display waiting time acquisition part 12 Quality estimation part 20 Web server 30 IP network 100 Drive apparatus 101 Recording medium 102 Auxiliary storage apparatus 103 Memory apparatus 104 CPU
105 Interface Device 106 Display Device 107 Input Device 121 QoE Acquisition Unit 122 QoE Average Value Calculation Unit 123 QoE Variation Calculation Unit 124 QoE Estimation Model Storage Unit 125 Web Browsing QoE Estimation Value Calculation Unit B Bus

Claims (8)

An acquisition unit that acquires a QoE (Quality of Experience) estimated value based on a display waiting time from when a display instruction for the Web page is generated until the Web page is displayed for each of a plurality of Web pages to be transitioned; ,
A first calculation unit that calculates an average value of a plurality of QoE estimation values acquired by the acquisition unit;
A second calculation unit that calculates variations in a plurality of QoE estimation values acquired by the acquisition unit;
A third calculating unit that calculates an estimated value of Web browsing QoE that is QoE after transitioning the plurality of Web pages based on the average value and the variation;
A web browsing quality estimation apparatus characterized by comprising: A QoE estimation model storage unit that stores an average value of a plurality of QoE estimation values, a variation of the plurality of QoE estimation values, and a QoE estimation formula and a coefficient representing a relationship between Web browsing QoE;
The third calculation unit inputs the average value calculated by the first calculation unit and the variation calculated by the second calculation unit to the QoE estimation formula to which the coefficient is applied. By calculating the estimated value of the Web browsing QoE,
The web browsing quality estimation apparatus according to claim 1. The second calculation unit calculates, as the variation, an absolute value average with respect to a difference between preceding and subsequent QoE estimation values, or an absolute value average of decrease values only when the QoE estimation value decreases. Item 3. The Web browsing quality estimation apparatus according to Item 1 or 2. An acquisition procedure for acquiring a QoE (Quality of Experience) estimate value for each of a plurality of Web pages to be transitioned based on a display waiting time from when a display instruction for the Web page is generated until the Web page is displayed; ,
A first calculation procedure for calculating an average value of a plurality of QoE estimation values acquired in the acquisition procedure;
A second calculation procedure for calculating variations of a plurality of QoE estimation values acquired in the acquisition procedure;
A third calculation procedure for calculating an estimated value of Web browsing QoE, which is QoE after transitioning the plurality of Web pages, based on the average value and the variation;
A Web browsing quality estimation method, wherein the computer executes The computer includes a QoE estimation model storage that stores an average value of a plurality of QoE estimation values, a variation in the plurality of QoE estimation values, and a QoE estimation expression and a coefficient representing a relationship between Web browsing QoE,
In the third calculation procedure, the average value calculated in the first calculation procedure and the variation calculated in the second calculation procedure are input to the QoE estimation formula to which the coefficient is applied. By calculating the estimated value of the Web browsing QoE,
The web browsing quality estimation method according to claim 4, wherein: The second calculation procedure calculates, as the variation, an absolute value average with respect to a difference between preceding and subsequent QoE estimated values, or an absolute value average of decreased values only when the QoE estimated value decreases. Item 6. The Web browsing quality estimation method according to Item 4 or 5.   The program for functioning a computer as each part of any one of Claims 1 thru | or 3. For each of a plurality of Web pages to be transitioned, a procedure for acquiring a QoE (Quality of Experience) estimated value based on a display waiting time from when a display instruction for the Web page is generated until the Web page is displayed;
A procedure for calculating an estimated value of Web browsing QoE, which is QoE after transitioning the plurality of Web pages, as a variation in which the average value of the QoE estimated values is corrected to be smaller as the variation in the estimated QoE value is larger.
A Web browsing quality estimation method, wherein the computer executes

Images