Timing calculation device, timing calculation method, and program

Abstract

In the prediction of a future event which shows, or might show, chaotic behavior, the purpose of the present disclosure is to clarify the timing at which at least two prediction values, which are to be compared, begin to diverge more than a prescribed amount or at least a prescribed amount, in the prediction of a future event which shows, or might show, chaotic behavior.　To that end, the present disclosure is a timing calculation device that includes: an input unit that receives input of a first initial value and a second initial value which are necessary in order to predict a future event which shows, or might show, chaotic behavior; a prediction value calculation unit that calculates a first prediction value for the event with respect to the first initial value, and also calculates a second prediction value for the event with respect to the second initial value; a divergence value calculation unit that calculates a divergence value based on the difference between the first prediction value and the second prediction value; a timing calculation unit that calculates a timing at which the divergence value becomes greater than a divergence threshold value, or a prescribed timing at which the divergence value is at least the divergence threshold value; and an output unit that outputs information relating to the prescribed timing.

Description

[Technical Field]

[0001] The present disclosure relates to a technique for calculating a time when a predicted value that deviates from a predetermined value can be adopted. [Background technology]

[0002] There are many forecasting techniques that use numerical calculations, etc. However, it is known that for phenomena that exhibit chaotic behavior, such as weather, even a slight difference in the initial values ​​input into the forecast model can cause large fluctuations in the forecast values.

[0003] To address this issue, for example, in weather forecasting, different initial values ​​are used, and each predicted value is derived as time-series data, and the range of fluctuation is clarified, demonstrating that the uncertainty of the prediction increases the further into the future, and then this is put to use in the real world (see non-patent document 1). [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Roberto Buizza, 2002, Chaos and weather prediction, European Center for Medium-Range Weather<https: / / www.ecmwf.int / file / 47912 / download?token=Qx-oYr4e> Summary of the Invention [Problem to be solved by the invention]

[0005] However, similar issues arise when predicting future events that exhibit or may exhibit chaotic behavior, not just in weather forecasting, if the initial values ​​cannot be set deterministically and uniquely. In many cases, the impact of differences in the initial values ​​on the predicted values ​​suddenly increases at a certain point in time. That is, there is relatively little difference for a while after the start of the forecast, but at some point the predicted values ​​begin to diverge suddenly. For example, if a forecasting model allows for a certain degree of range in the predicted values, it is important to understand when the predicted values ​​begin to diverge suddenly. This is because it makes it possible to clearly indicate information that can be used to determine the temporal scope of use of the forecasting model, for example.

[0006] The present invention has been made in consideration of the above points, and aims to clarify the time when at least two predicted values ​​to be compared will become larger than a predetermined value or will begin to diverge by more than a predetermined value when predicting future events that exhibit or may exhibit chaotic behavior. [Means for solving the problem]

[0007] In order to solve the above problem, the invention according to claim 1 provides a method for predicting a future event that exhibits or may exhibit chaotic behavior, comprising: an input unit for inputting a first initial value and a second initial value required for predicting a future event that exhibits or may exhibit chaotic behavior; Using a prediction model that outputs a predicted value for an input of an initial value, a predicted value calculation unit that calculates a first predicted value of the event for the first initial value and a second predicted value of the event for the second initial value; a deviation value calculation unit that calculates a deviation value based on a difference between the first predicted value and the second predicted value; and a time when the deviation value becomes larger than a deviation threshold or more than the time when the deviation value becomes larger than a deviation threshold. begin to diverge The timing calculation device includes a timing calculation unit that calculates a predetermined timing, and an output unit that outputs information relating to the predetermined timing. [Effects of the Invention]

[0008] As described above, the present invention has the advantage of being able to clarify the time when at least two predicted values ​​to be compared will become larger than a predetermined value or will begin to diverge by more than a predetermined value when predicting future events that exhibit or may exhibit chaotic behavior. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a schematic diagram of a communication system. [Figure 2] FIG. 2 is an electrical hardware configuration diagram of the timing calculation device. [Figure 3] FIG. 2 is a diagram illustrating an electrical hardware configuration of a communication terminal. [Figure 4] FIG. 2 is a functional configuration diagram of a timing calculation device according to an embodiment. [Figure 5] 10 is a flowchart showing a process for calculating the time when deviation begins due to a difference in initial values. [Figure 6] FIG. 10 is a diagram showing the time-series locus of predicted values ​​output when different initial values ​​are input to a prediction model. [Figure 7] FIG. 10 is a diagram showing an example of plotting an initial value and n, which mainly indicates the passage of time. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0011] [System configuration of the embodiment] First, an outline of the configuration of a communication system according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a schematic diagram of a communication system according to an embodiment of the present invention.

[0012] As shown in Fig. 1, the communication system 1 of this embodiment is constructed by a timing calculation device 3 and a communication terminal 5. The communication terminal 5 is managed and used by a user Y. The user refers to the output result of the timing calculation device and determines up to what future time (including "time point") the predicted value should be adopted.

[0013] Furthermore, the timing calculation device 3 and the communication terminal 5 can communicate with each other via a communication network 100 such as the Internet. The connection form of the communication network 100 may be either wireless or wired.

[0014] The time calculation device 3 is configured by one or more computers. When the time calculation device 3 is configured by multiple computers, it may be referred to as a "time calculation device" or a "time calculation system."

[0015] The timing calculation device 3 predicts future events that exhibit or may exhibit chaotic behavior, and calculates the time when at least two predicted values ​​to be compared will become larger than a predetermined value or will begin to diverge by more than a predetermined value. Events that exhibit or may exhibit chaotic behavior include weather predictions, stock price predictions, etc.

[0016] The communication terminal 5 is a computer, and a notebook computer is shown as an example in Fig. 1. In Fig. 1, a user Y operates the communication terminal 5. Note that the timing calculation device 3 may perform processing independently without using the communication terminal 5.

[0017] [Hardware configuration] <Hardware configuration of timing calculation device> Next, the electrical hardware configuration of the timing calculation device 3 will be described with reference to Fig. 2. Fig. 2 is a diagram showing the electrical hardware configuration of the timing calculation device.

[0018] As shown in FIG. 2, the timing calculation device 3 is a computer and includes a CPU (Central Processing Unit) 301, a ROM (Read Only Memory) 302, a RAM (Random Access Memory) 303, an SSD (Solid State Drive) 304, an external device connection I / F (Interface) 305, a network I / F 306, a media I / F 309, and a bus line 310.

[0019] Of these, the CPU 301 controls the overall operation of the timing calculation device 3. The ROM 302 stores programs such as an IPL (Initial Program Loader) used to drive the CPU 301. The RAM 303 is used as a work area for the CPU 301.

[0020] The SSD 304 reads or writes various data under the control of the CPU 301. Note that instead of the SDD 304, a HDD (Hard Disk Drive) may be used.

[0021] The external device connection I / F 305 is an interface for connecting various external devices, such as a display, a speaker, a keyboard, a mouse, a USB (Universal Serial Bus) memory, and a printer.

[0022] The network I / F 306 is an interface for performing data communication via the communication network 100 .

[0023] The media I / F 309 controls reading and writing (storing) of data from and to a recording medium 309m such as a flash memory, etc. The recording medium 309m includes a DVD (Digital Versatile Disc) and a Blu-ray Disc (registered trademark).

[0024] The bus line 310 is an address bus, a data bus, etc. for electrically connecting the components such as the CPU 301 shown in FIG.

[0025] <Hardware configuration of communication terminal> Next, the electrical hardware configuration of the communication terminal 5 will be described with reference to Fig. 3. Fig. 3 is a diagram showing the electrical hardware configuration of the communication terminal.

[0026] As shown in FIG. 3, the communication terminal 5 is a computer and includes a CPU 501, a ROM 502, a RAM 503, an SSD 504, an external device connection I / F (Interface) 505, a network I / F 506, a display 507, a pointing device 508, a media I / F 509, and a bus line 510.

[0027] Of these, the CPU 501 controls the overall operation of the communication terminal 5. The ROM 502 stores programs such as IPL used to drive the CPU 501. The RAM 503 is used as a work area for the CPU 501.

[0028] The SSD 504 reads or writes various data under the control of the CPU 501. Note that instead of the SSD 504, an HDD (Hard Disk Drive) may be used.

[0029] The external device connection I / F 505 is an interface for connecting various external devices, such as a display, a speaker, a keyboard, a mouse, a USB memory, and a printer.

[0030] The network I / F 506 is an interface for performing data communication via the communication network 100 .

[0031] The display 507 is a type of display means such as a liquid crystal display or organic EL (Electro Luminescence) display that displays various images.

[0032] The pointing device 508 is a type of input means for selecting and executing various instructions, selecting a processing target, moving a cursor, etc. If user Y uses a keyboard, the function of the pointing device 508 may be turned off.

[0033] The media I / F 509 controls reading and writing (storing) of data from and to a recording medium 509m such as a flash memory, etc. The recording medium 509m includes DVDs, Blu-ray Discs (registered trademarks), etc.

[0034] The bus line 510 is an address bus, a data bus, etc. for electrically connecting the components such as the CPU 501 shown in FIG.

[0035] [Functional configuration of timing calculation device] Next, the functional configuration of the timing calculation device will be described with reference to Fig. 4. Fig. 4 is a functional configuration diagram of the timing calculation device in the embodiment.

[0036] 4, the timing calculation device 3 has an input unit 31, a predicted value calculation unit 32, a deviation value calculation unit 33, a timing calculation unit 34, a formulation unit 35, and an output unit 39. Each of these units has a function realized by an instruction from the CPU 301 in FIG. 2 based on a program.

[0037] <Prediction model> A prediction model 30 used to predict future events that exhibit or may exhibit chaotic behavior is constructed in the RAM 303 or HD 304 in Fig. 2. The format and creation method of the prediction model 30 are not important.

[0038] Here, a logistic map is used as the prediction model 30 for the purpose of explanation, and the definition formula is shown in (Formula 1).

[0039]

number

[0040] For the sake of simplicity, the parameter r is used as one parameter here, but it may be an equation containing multiple parameters. Also, the prediction model 30 is shown as a single equation, but it may be in the form of a machine learning model such as a neural network.

[0041] <Each function configuration> Next, each functional configuration of the timing calculation device will be described with reference to FIGS.

[0042] The input unit 31 receives input of setting values ​​such as at least a first initial value and a second initial value for the prediction model from the user Y via the communication terminal 5 and the network I / F 306.

[0043] The predicted value calculation unit 32 uses the prediction model 30 to calculate a first predicted value of a future event for a first initial value, and calculates a second predicted value of a future event for a second initial value.

[0044] The deviation value calculation unit 33 calculates a deviation value from the difference between the first predicted value and the second predicted value.

[0045] The timing calculation unit 34 calculates the timing when the deviation value becomes larger than the deviation threshold value or a predetermined timing when the deviation value becomes equal to or greater than the deviation threshold value.

[0046] The formulation unit 35 formulates the variability in the timing at which the predicted values ​​start to diverge. That is, the formulation unit 35 formulates the relationship between the changes in the first initial value and the second initial value and the elapsed time related information T that indicates the elapsed time into the future or information related to the elapsed time.

[0047] The output unit 38 outputs information relating to the predetermined time calculated by the time calculation unit 34. The output unit 38 also outputs information on the result formulated by the formulation unit 35. Examples of the output method include transmitting data of the output result to the communication terminal 5 via the network I / F 506, displaying it on an external display via the external device connection I / F 505, or printing it by a printer or the like via the external device connection I / F 505.

[0048] [Processing or Operation of the Embodiment] Next, the processing or operation of this embodiment will be described in detail with reference to Figures 5 to 7. Figure 5 is a flowchart showing the processing for calculating the timing at which deviation begins due to the difference in initial values.

[0049] S11: The input unit 31 inputs various setting values ​​(first initial value, second initial value, etc.) required for processing from the communication terminal 5 or the like.

[0050] S12: The predicted value calculation unit 32 inputs each initial value input by the input unit 31 into the prediction model 30 and outputs each predicted value, thereby calculating each predicted value. Here, the processing of step S12 will be described in detail with reference to Fig. 6. Fig. 6 is a diagram showing the time-series trajectory of predicted values ​​output when different initial values ​​are input to the prediction model.

[0051] Figure 6 shows the trajectory of the prediction model when r = a and there is a very small difference in the initial value x0 (n = 0) in (Equation 1). In Figure 6, the horizontal axis represents the value of n in (Equation 1), and the vertical axis represents the value of x in (Equation 1). The solid line represents the trajectory when the prediction model 30 is given an initial value x0, and the dashed line represents the trajectory when a very small value α is added to the initial value x0. The two trajectories move almost identically for a while, but at a certain point, they rapidly diverge. The setting range of the initial value is determined by the target phenomenon, or a user Y arbitrarily sets the initial value range he or she wants to check. A representative value of x0 is selected from that range. This representative value is designated X. The representative value X may be automatically set as the median value of the setting range, or it may be set manually as appropriate. The representative value is an example of a first initial value, and an initial value other than the representative value is an example of a second initial value.

[0052] S13: The deviation value calculation unit 33 calculates a deviation value using the predicted value calculated by the predicted value calculation unit 32. The deviation value is calculated by subtracting the initial value x calculated using the representative value X (first initial value) of the initial values ​​and each initial value other than the representative value X (second initial value). n Alternatively, if there are multiple second initial values, the multiple initial values ​​x n The deviation value may be calculated by other calculation methods, such as the average or sum of the deviation values ​​in

[0053] S14: The timing calculation unit 34 determines whether the deviation value calculated by the deviation value calculation unit 33 satisfies the deviation threshold condition. In this case, the timing calculation unit 34 compares the deviation value calculated by the deviation value calculation unit 33 with a preset deviation threshold, and determines that a deviation of a predetermined state has begun if the deviation value is greater than the deviation threshold (or if the deviation value is equal to or greater than the deviation threshold). The deviation threshold may be set automatically by some method or manually as appropriate.

[0054] The timing calculation unit 34 repeats the calculation of step S14 within the range of possible values ​​of the parameter r. As a result, a set of values ​​of n at which the predicted values ​​begin to diverge is found within the range of possible values ​​of the parameter r and the initial value x0. FIG. 7 is a diagram showing an example of a plot of the initial value and n, which mainly represents the passage of time. FIG. 7 shows an example of a plot of n in (Equation 1). The horizontal axis represents x0, and the vertical axis represents n.

[0055] If the timing calculation unit 34 determines that the deviation value does not satisfy the deviation threshold condition, the process proceeds to step S15, and if the timing calculation unit 34 determines that the deviation value satisfies the deviation threshold condition, the process proceeds to step S16.

[0056] S15: The timing calculation unit 34 determines whether or not to change the various setting values. If the timing calculation unit 34 determines that the various setting values ​​should be changed, the process returns to step S11, the input unit 31 changes the various setting values, and the subsequent processing is performed. On the other hand, if the timing calculation unit 34 determines that the various setting values ​​should not be changed, the processing shown in FIG. 5 ends.

[0057] S16: The output unit 39 outputs information such as a graph (see FIG. 7) showing the relationship between the change in the initial value and the elapsed time related information T. This allows the user Y to visually check the information such as the graph, thereby deepening the user Y's understanding. Note that this step S16 may be omitted.

[0058] S17: The formulation unit 35 formulates the relationship between the change in the initial value and the elapsed time related information T. Specifically, the formulation unit 35 formulates the relationship between the parameter r, the initial value x0, and n. Any method or function may be used for the formulation. A multidimensional simultaneous distribution with respect to n may be used, or a two-dimensional distribution may be obtained for each combination with n, such as the simultaneous distribution of the initial value x0 and n.

[0059] S18: The output unit 39 outputs the results formulated by the formulation unit 35.

[0060] [Effects of the embodiment] As described above, this embodiment has the advantage of being able to clarify the time when at least two predicted values ​​to be compared will become larger than a predetermined value or will begin to diverge by more than a predetermined value when predicting future events that exhibit or may exhibit chaotic behavior. This makes it easier for user Y to determine up to what time point the predicted values ​​should be used.

[0061] 〔supplement〕 The present invention is not limited to the above-described embodiment, and may have the following configurations or processes (operations). (1) The timing calculation device 3 can be realized by a computer and a program, but this program can also be recorded on a (non-transitory) recording medium or provided via the communication network 100. (2) In the communication between the timing calculation device 3 and the communication terminal 5, other devices (such as a server or a router) may relay data. For example, for the sake of simplicity, this specification describes that the input unit 31 of the timing calculation device 3 transmits data to the communication terminal 5, but this transmission process also includes cases where other devices relay data. (4) In the above embodiment, a notebook computer is shown as an example of a communication terminal 5, but this is not limited to this and may be, for example, a desktop computer, a tablet terminal, a smartphone, a smartwatch, a car navigation device, a refrigerator, a microwave oven, etc. (5) Each CPU 301, 501 may be a single CPU or may be multiple CPUs. [Explanation of symbols]

[0062] 1. Communication Systems 3 Timing calculation device 5. Communication terminals 30 Predictive Models 31 Input section 32 Prediction value calculation section 33 Deviation value calculation unit 34 Timing calculation part 35 Formulation part 36 Output section

Claims

1. an input unit for inputting a first initial value and a second initial value required to predict a future event that exhibits or may exhibit chaotic behavior; a prediction value calculation unit that calculates a first prediction value of the event for the first initial value and a second prediction value of the event for the second initial value using a prediction model that outputs a prediction value for an input of an initial value; a deviation value calculation unit that calculates a deviation value based on a difference between the first predicted value and the second predicted value; a timing calculation unit that calculates a predetermined time when the deviation value becomes larger than the deviation threshold value or when the deviation value begins to deviate beyond the deviation threshold value; an output unit that outputs information related to the predetermined time period; A timing calculation device having the above.

2. 2. The timing calculation device according to claim 1, a formulation unit that formulates the first predicted value and the second predicted value calculated by the predicted value calculation unit in accordance with the predetermined time period calculated by the time period calculation unit.

3. A timing calculation device as described in claim 1 or 2, wherein the prediction model uses a logistic map.

4. The timing calculation device according to claim 3 , wherein the prediction model is a machine learning model using a neural network.

5. The timing calculation device according to claim 1 or 2, wherein the event is a weather forecast or a stock price forecast.

6. A computer comprising: inputting a first initial value and a second initial value required to predict future events that exhibit or are likely to exhibit chaotic behavior; calculating a first predicted value of the event for the first initial value and a second predicted value of the event for the second initial value using a prediction model that outputs a predicted value for an input of an initial value; calculating a deviation value based on a difference between the first predicted value and the second predicted value; calculating a time when the deviation value becomes larger than the deviation threshold value or a predetermined time when the deviation value begins to deviate beyond the deviation threshold value; outputting information relating to the predetermined time period; Time calculation method.

7. A program causing a computer to execute the method according to claim 6.

Images