CN102819768A - Method and system for analyzing passenger flow data - Google Patents
Method and system for analyzing passenger flow data Download PDFInfo
- Publication number
- CN102819768A CN102819768A CN2011103483166A CN201110348316A CN102819768A CN 102819768 A CN102819768 A CN 102819768A CN 2011103483166 A CN2011103483166 A CN 2011103483166A CN 201110348316 A CN201110348316 A CN 201110348316A CN 102819768 A CN102819768 A CN 102819768A
- Authority
- CN
- China
- Prior art keywords
- passenger flow
- flow data
- estimated
- historical
- msub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 36
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 238000007405 data analysis Methods 0.000 claims description 24
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012706 support-vector machine Methods 0.000 abstract 1
- 239000011159 matrix material Substances 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000013215 result calculation Methods 0.000 description 2
- 238000010187 selection method Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for analyzing passenger flow data. The method comprises the steps of collecting historical passenger flow data including an estimated variable, building an estimated module according to the historical passenger flow data, and estimating the passenger flow data according to the estimated variable and the estimated module. According to the invention, a sample set for an LS-SVM (Least Squares-Support Vector Machine) model can be constructed through collecting the historical passenger flow data; the sample set is subjected to back training by using an LS-SVM algorithm; relevant parameters of regression function are calculated and the regression function is constructed; and then a preset estimated variable value is input to the regression function; corresponding forecast passenger flow data is calculated so as to obtain more correct estimated data.
Description
Technical Field
The invention relates to the field of data analysis, in particular to a passenger flow data analysis method and system.
Background
In traditional industries, such as the catering industry and the like, accurate analysis of passenger flow data is not realized, and the accuracy is difficult to guarantee through artificial subjective judgment of a manager.
With the development of science and technology, information technology is introduced into the traditional industry, passenger flow data is accurately analyzed, help is provided for the decision of a manager, and the method is a necessary trend for improving the management level of the traditional industry. Therefore, it is an urgent need to provide a method or system for accurately estimating passenger flow data.
Disclosure of Invention
The invention mainly aims to provide a passenger flow data analysis method, which can realize the estimation of passenger flow data and improve the accuracy of passenger flow data estimation.
The invention provides a passenger flow data analysis method, which comprises the following steps:
collecting historical passenger flow data containing pre-estimated variables;
establishing a pre-estimation model according to the historical passenger flow data;
and estimating passenger flow data according to the estimated variables and the estimated model.
Preferably, the method further comprises:
and setting an estimated variable of the passenger flow data.
Preferably, the predictor variables include weather conditions and holiday conditions.
Preferably, the step of estimating the passenger flow data according to the estimated variables and the estimated model specifically includes:
performing regression training on the historical data by adopting an LS _ SVM algorithm, calculating relevant parameters of a regression function, and constructing the regression function;
and putting the estimated variables into the regression function to obtain the estimated passenger flow data.
Preferably, the step of calculating the regression function correlation parameter and constructing the regression function specifically includes:
calculating to obtain corresponding parameters b and alphaiAnd constructing a prediction function:
the invention also provides a system for passenger flow data analysis, which comprises:
the data acquisition unit is used for acquiring historical passenger flow data containing the pre-estimated variables;
the model establishing unit is used for establishing a pre-estimation model according to the historical passenger flow data;
and the estimation calculation unit is used for estimating the passenger flow data according to the estimation variable and the estimation model.
Preferably, the system further comprises:
and the variable setting unit is used for setting the estimated variable of the passenger flow data.
Preferably, the predictor variables include weather conditions and holiday conditions.
Preferably, the pre-estimation calculating unit specifically includes:
the function building module is used for carrying out regression training on the historical data by adopting an LS _ SVM algorithm, calculating relevant parameters of a regression function and building the regression function;
and the data acquisition module is used for putting the pre-estimated variables into the regression function to acquire the estimated passenger flow data.
Preferably, the function building module is specifically configured to:
calculating to obtain corresponding parameters b and alphaiAnd constructing a prediction function:
the invention can construct a sample set of an LS _ SVM model by acquiring historical passenger flow data, performs regression training on the sample set by adopting an LS _ SVM algorithm, calculates the relevant parameters of a regression function to construct a regression function, inputs preset estimation variable values into the regression function, and calculates to obtain corresponding predicted passenger flow data, thereby obtaining more accurate estimation data.
Drawings
FIG. 1 is a flow chart illustrating steps in an embodiment of a method for passenger flow data analysis according to the present invention;
FIG. 2 is a flow chart illustrating steps in another embodiment of a method for passenger flow data analysis according to the present invention;
FIG. 3 is a flow chart illustrating another step of another embodiment of the method for passenger flow data analysis of the present invention;
FIG. 4 is a schematic block diagram of an embodiment of a system for analyzing passenger flow data according to the present invention;
FIG. 5 is a schematic diagram of another embodiment of a system for passenger flow data analysis according to the present invention;
FIG. 6 is a schematic diagram of a pre-estimation computing unit in another embodiment of the system for passenger flow data analysis according to the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, an embodiment of a method for analyzing traffic data is provided. The method can comprise the following steps:
step S10, collecting historical passenger flow data containing the estimated variables;
step S11, establishing an estimation model according to the historical passenger flow data;
and step S12, estimating the passenger flow data according to the estimated variables and the estimated model.
The estimation of the passenger flow (volume) data in this embodiment is calculated and obtained by a least squares support vector regression (LS _ SVM) algorithm based on the historical data of the passenger flow.
Firstly, historical passenger flow data containing an estimated variable is required to be obtained, and the historical passenger flow data is mainly used for constructing a sample set of an LS _ SVM model. The estimated variables mainly refer to variable factors influencing passenger flow. Then, a pre-estimation model (which can be a regression function) is established according to the historical passenger flow data, and corresponding passenger flow data is obtained according to the pre-estimation variables and the pre-estimation model. The estimation mode based on historical passenger flow data by using the LS _ SVM algorithm can enable estimation to be more accurate.
Referring to fig. 2, in another embodiment of the present invention, before the step S10, the method further includes:
and S100, setting an estimated variable of the passenger flow data.
Since the estimated variables are main variable factors influencing passenger flow, the corresponding estimated variables are required to be set as estimation references before estimation. Meanwhile, the collected historical passenger flow data at least needs to include the passenger flow data and the corresponding pre-estimated variables to be complete. The predictive variable settings may also refer to historical passenger flow data.
For example, the passenger flow data of the XX catering enterprise in the schedule of 2010/08/09-2010/09/10 days and 33 days (see Table 1) can be collected, and the passenger flow data of 2010/09/11-2010/09/14 days and 5 days can be estimated. The projected variables may include weather conditions, holiday conditions, etc. The historical passenger flow data may include weather conditions, holiday conditions, and corresponding passenger flow data, among others.
| Date | Week | Passenger flow volume | Weather (weather) |
| 8 month and 9 days | A | 251 | All-weather |
| 8 month and 10 days | II | 234 | All-weather |
| 8 month and 11 days | III | 255 | All-weather |
| ... | ... | ... | ... |
| 9 month and 8 days | III | 206 | All-weather |
| 9 months and 9 days | Fourthly | 252 | All-weather |
| 9 month and 10 days | Five of them | 274 | All-weather |
TABLE 1
The data types of the estimated variables are mainly three, including enumeration type, Boolean type and numerical type, and can be realized through a left tree and a right table structure. Creating an estimated variable under the left tree root node, and when the estimated variable is a numerical type and a Boolean type, the right table does not allow to create detailed data; when the estimated variables are enumerated, the right table creates corresponding enumerated values, and a characteristic value is created for each enumerated value, and the setting trend and the distance of the characteristic value must be consistent with the trend of influencing the result due to the fact that the characteristic value participates in the result calculation.
Because the operation of the catering industry has certain continuity, the analysis store generally takes the passenger flow data in the previous period as the influence factor for predicting the passenger flow data in the next period. Analyzing the above example, creating estimated variables, which are weather conditions and holiday conditions, respectively, and correspondingly creating enumerated values as shown in table 2:
TABLE 2
The collected historical passenger flow data is mainly used for constructing a sample set of the LS _ SVM model. Each sample data in the sample set may include projected variables, passenger flow volume, etc. for the current day. For convenience in operation, the historical passenger flow data can be recorded in a date-based data recording mode and a variable-based data recording mode.
The historical passenger flow data in the above example comprises passenger flow information of 2010/08/12-2010/09/10 in the first three days, daily weather conditions, holiday conditions and the current day, and a training sample set is constructed according to the historical passenger flow data as shown in table 3:
TABLE 3
The parameters involved in establishing the predictive model may include: training sample set number, sample collection days, maximum prediction days, prediction variable value and the like. The number of the training sample sets is the number of the sample sets participating in training; the number of the sample collection days is the number of days with the historical passenger flow as an influence factor; the maximum forecast days are the most days which can be forecasted by adopting the current forecast model; the estimated variable value is a characteristic value corresponding to the influence factor involved in calculation by adopting the current estimated model.
The structure of the prediction model constructed according to the above example analysis is shown in table 4:
TABLE 4
Referring to fig. 3, step S12 further includes:
step S121, performing regression training on the historical data by adopting an LS _ SVM algorithm, calculating relevant parameters of a regression function, and constructing the regression function;
and S122, putting the estimated variables into the regression function to obtain the estimated passenger flow data.
Based on the above estimatePerforming regression training on the sample set by adopting an LS _ SVM algorithm, and calculating relevant parameters of a regression function to construct a regression function; then, the forecast variable values (weather characteristic value, date characteristic value and corresponding historical passenger flow) corresponding to the forecast dates are input to obtain forecast passenger flow data. Performing estimation calculation according to an LS _ SVM algorithm, firstly determining a training set and a penalty factor, wherein in the example, the penalty factor C is 0.01 according to repeated experiments; then, an omega matrix is calculated according to the training set, regression training is carried out by utilizing historical passenger flow data, and corresponding parameters b and alpha are obtainediAnd constructing a prediction function:and then, a prediction value is obtained according to the prediction function (1). A list of predicted values can thus be obtained, as shown in table 5:
| business organization | Date | Forecast passenger flow | Actual passenger flow volume | Relative error |
| XX dinner book department | 2010-09-11 | 316 | 285 | 0.109 |
| XX dinner book department | 2010-09-12 | 316 | 287 | 0.101 |
| XX dinner book department | 2010-09-13 | 229 | 206 | 0.112 |
| XX dinner book department | 2010-09-14 | 229 | 221 | 0.036 |
| XX dinner book department | 2010-09-15 | 229 | 234 | 0.021 |
| Average relative error | 0.076 |
TABLE 5
As can be seen from the above Table 5, the comparison between the estimated passenger flow and the actual passenger flow basically controls the relative error to be about 10%, and the requirements of the industry can be met.
The LS _ SVM regression training is specifically implemented as follows:
firstly, determining variables required for constructing a regression equation, namely influence factors related to daily passenger capacity; such as date characteristic values and weather characteristic values; determining a numerical value corresponding to each value of enumeration type/Boolean type in the influence factors; selecting a penalty factor C; and selecting a training set, wherein the selection method of the training set, the number of samples of the training set, the input of each pair of strain values in the training set and the like are included.
Then, constructing a prediction function; 1. omega is obtainedijFrom the formula
And calculating to obtain the data, wherein, <math>
<mrow>
<msub>
<mi>δ</mi>
<mi>ij</mi>
</msub>
<mo>=</mo>
<mfenced open='{' close=''>
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mi>i</mi>
<mo>=</mo>
<mi>j</mi>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mi>i</mi>
<mo>≠</mo>
<mi>j</mi>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>,</mo>
</mrow>
</math> <math>
<mrow>
<msup>
<mi>g</mi>
<mi>T</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mi>g</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>j</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mi>T</mi>
</msubsup>
<mo>·</mo>
<msub>
<mi>x</mi>
<mi>j</mi>
</msub>
<mo>;</mo>
</mrow>
</math>
2. constructing omega matrix and inverting omega matrix-1；
3. The value of b can be obtained according to the formula b ═ ITΩ-1I)-1ITΩ-1y is calculated; the method specifically comprises the following steps: (1) to obtain IT·Ω-1(ii) a (2) To obtain IT·Ω-1I; (3) the above-mentioned value (2) is obtained as a numerical value, and the reciprocal (I) is obtainedT·Ω-1·I)-1(ii) a (4) Comparing the result of (3) with ITMultiplying; (5) comparing the result in (4) with omega-1Multiplying; (6) multiplying the result in (5) by y, thereby obtaining a b value;
4. finding alphaiValue can be given by the formula α ═ Ω-1(y-b I) performing a calculation; the method specifically comprises the following steps: (1) b, calculating b I; (2) calculating y-b I; (3) will omega-1Multiplying the result of the above (2) to obtain alphaiA value;
5. further, the above b value and alpha are measurediValue substitution prediction functionCalculating a predicted value; wherein,
i is a natural number from 1 to N.
Assume that the variables are 3, sample 1 (corresponding to matrix x)1) Respectively as follows: a1, a2, a 3; sample 2 (correspondence matrix x)2) Respectively as follows: b1, b2, b 3; then And is <math>
<mrow>
<msubsup>
<mi>x</mi>
<mn>1</mn>
<mi>T</mi>
</msubsup>
<mo>·</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>=</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>a</mi>
<mrow>
<mn>1</mn>
<mo>,</mo>
</mrow>
</msub>
<msub>
<mi>a</mi>
<mn>2</mn>
</msub>
<mo>,</mo>
<msub>
<mi>a</mi>
<mn>3</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>·</mo>
<mfenced open='(' close=')'>
<mtable>
<mtr>
<mtd>
<msub>
<mi>a</mi>
<mn>1</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>a</mi>
<mn>2</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>a</mi>
<mn>3</mn>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
</math> <math>
<mrow>
<msubsup>
<mi>x</mi>
<mn>1</mn>
<mi>T</mi>
</msubsup>
<mo>·</mo>
<msub>
<mi>x</mi>
<mn>2</mn>
</msub>
<mo>=</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>a</mi>
<mrow>
<mn>1</mn>
<mo>,</mo>
</mrow>
</msub>
<msub>
<mi>a</mi>
<mn>2</mn>
</msub>
<mo>,</mo>
<msub>
<mi>a</mi>
<mn>3</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>·</mo>
<mfenced open='(' close=')'>
<mtable>
<mtr>
<mtd>
<msub>
<mi>b</mi>
<mn>1</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>b</mi>
<mn>2</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>b</mi>
<mn>3</mn>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
</math> When there are n samples, and so on.
The passenger flow data analysis method can construct a sample set of an LS _ SVM model by collecting historical passenger flow data, carries out regression training on the sample set by adopting an LS _ SVM algorithm, calculates the relevant parameters of a regression function to construct a regression function, inputs preset estimation variable values into the regression function, and calculates to obtain corresponding predicted passenger flow data, thereby obtaining more accurate estimation data.
Referring to fig. 4, an embodiment of a system 20 for passenger flow data analysis of the present invention is presented. The system 20 may include: a data acquisition unit 21, a model establishing unit 22 and an estimation calculating unit 23; the data acquisition unit 21 is used for acquiring historical passenger flow data containing pre-estimated variables; the model establishing unit 22 is configured to establish an estimation model according to the historical passenger flow data; the estimation calculating unit 23 is used for estimating the passenger flow data according to the estimation variables and the estimation model.
The estimation of the passenger flow (volume) data in this embodiment is calculated and obtained by a least squares support vector regression (LS _ SVM) algorithm based on the historical data of the passenger flow.
Firstly, historical passenger flow data containing an estimated variable is acquired through a data acquisition unit 21, and the historical passenger flow data is mainly used for constructing a sample set of an LS _ SVM model. The estimated variables mainly refer to variable factors influencing passenger flow. Then, the model building unit 22 builds a pre-estimation model (which may be a regression function) according to the historical passenger flow data, and the pre-estimation calculating unit 23 obtains corresponding passenger flow data according to the pre-estimation variables and the pre-estimation model. The estimation mode based on historical passenger flow data by using the LS _ SVM algorithm can enable estimation to be more accurate.
Referring to fig. 5, the system 20 further includes: and the variable setting unit 24 is used for setting the estimated variable of the passenger flow data.
Since the estimated variables are main variable factors influencing passenger flow, the corresponding estimated variables are required to be set as estimation references before estimation. Meanwhile, the collected historical passenger flow data at least needs to include the passenger flow data and the corresponding pre-estimated variables to be complete. The predictive variable settings may also refer to historical passenger flow data.
For example, the passenger flow data of the XX catering enterprise in the schedule of 2010/08/09-2010/09/10 days and 33 days (see Table 1) can be collected, and the passenger flow data of 2010/09/11-2010/09/14 days and 5 days can be estimated. The projected variables may include weather conditions, holiday conditions, etc. The historical passenger flow data may include weather conditions, holiday conditions, and corresponding passenger flow data, among others.
The data types of the estimated variables are mainly three, including enumeration type, Boolean type and numerical type, and can be realized through a left tree and a right table structure. Creating an estimated variable under the left tree root node, and when the estimated variable is a numerical type and a Boolean type, the right table does not allow to create detailed data; when the estimated variables are enumerated, the right table creates corresponding enumerated values, and a characteristic value is created for each enumerated value, and the setting trend and the distance of the characteristic value must be consistent with the trend of influencing the result due to the fact that the characteristic value participates in the result calculation.
Because the operation of the catering industry has certain continuity, the analysis store generally takes the passenger flow data in the previous period as the influence factor for predicting the passenger flow data in the next period. Analyzing the above example, creating estimated variables, which are weather conditions and holiday conditions, respectively, and correspondingly creating enumerated values as shown in table 2.
The collected historical passenger flow data is mainly used for constructing a sample set of the LS _ SVM model. Each sample data in the sample set may include projected variables, passenger flow volume, etc. for the current day. For convenience in operation, the historical passenger flow data can be recorded in a date-based data recording mode and a variable-based data recording mode.
The historical passenger flow data in the above example comprises passenger flow information of 2010/08/12-2010/09/10 in the first three days, daily weather conditions, holiday conditions and current day, and a training sample set is constructed according to the historical passenger flow data as shown in table 3.
The parameters involved in establishing the predictive model may include: training sample set number, sample collection days, maximum prediction days, prediction variable value and the like. The number of the training sample sets is the number of the sample sets participating in training; the number of the sample collection days is the number of days with the historical passenger flow as an influence factor; the maximum forecast days are the most days which can be forecasted by adopting the current forecast model; the estimated variable value is a characteristic value corresponding to the influence factor involved in calculation by adopting the current estimated model.
The structure of the prediction model constructed according to the above example analysis is shown in table 4.
Referring to fig. 6, the estimation calculating unit 23 specifically includes: a function construction module 231 and a data acquisition module 232; the function constructing module 231 is used for performing regression training on the historical data by adopting an LS _ SVM algorithm and calculating regressionFunction-related parameters, constructing a regression function; the data obtaining module 232 is configured to put the estimated variable into the regression function to obtain estimated passenger flow data. The function building module 231 is specifically configured to: calculating to obtain corresponding parameters b and alphaiAnd constructing a prediction function: <math>
<mrow>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mi>Σ</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mi>n</mi>
</munderover>
<mrow>
<mo>(</mo>
<msub>
<mi>α</mi>
<mi>i</mi>
</msub>
<mo>*</mo>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mi>T</mi>
</msubsup>
<mo>)</mo>
</mrow>
<mo>*</mo>
<mi>x</mi>
<mo>+</mo>
<mi>b</mi>
<mo>.</mo>
</mrow>
</math>
the function building module 231 may perform regression training on the sample set by using an LS _ SVM algorithm according to the pre-estimated model, and calculate regression function correlation parameters to build a regression function; the data obtaining module 232 obtains the predicted passenger flow data according to the predicted variable values (weather characteristic value, date characteristic value and corresponding historical passenger flow) corresponding to the input predicted dates. Performing estimation calculation according to an LS _ SVM algorithm, firstly determining a training set and a penalty factor, wherein in the example, the penalty factor C is 0.01 according to repeated experiments; then, an omega matrix is calculated according to the training set, regression training is carried out by utilizing historical passenger flow data, and corresponding parameters b and alpha are obtainediAnd constructing a prediction function:and then, a prediction value is obtained according to the prediction function (1). A list of predicted values can thus be obtained, as shown in table 5.
As can be seen from the above Table 5, the comparison between the estimated passenger flow and the actual passenger flow basically controls the relative error to be about 10%, and the requirements of the industry can be met.
The LS _ SVM regression training is specifically implemented as follows:
firstly, determining variables required for constructing a regression equation, namely influence factors related to daily passenger capacity; such as date characteristic values and weather characteristic values; determining a numerical value corresponding to each value of enumeration type/Boolean type in the influence factors; selecting a penalty factor C; and selecting a training set, wherein the selection method of the training set, the number of samples of the training set, the input of each pair of strain values in the training set and the like are included.
Then, constructing a prediction function; 1. omega is obtainedijFrom the formula
And calculating to obtain the data, wherein, <math>
<mrow>
<msub>
<mi>δ</mi>
<mi>ij</mi>
</msub>
<mo>=</mo>
<mfenced open='{' close=''>
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mi>i</mi>
<mo>=</mo>
<mi>j</mi>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mi>i</mi>
<mo>≠</mo>
<mi>j</mi>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>,</mo>
</mrow>
</math> <math>
<mrow>
<msup>
<mi>g</mi>
<mi>T</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mi>g</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>j</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mi>T</mi>
</msubsup>
<mo>·</mo>
<msub>
<mi>x</mi>
<mi>j</mi>
</msub>
<mo>;</mo>
</mrow>
</math>
2. constructing omega matrix and inverting omega matrix-1；
3. The value of b can be obtained according to the formula b ═ ITΩ-1I)-1ITΩ-1y is calculated; the method specifically comprises the following steps: (1) to obtain IT·Ω-1(ii) a (2) To obtain IT·Ω-1I; (3) the above-mentioned value (2) is obtained as a numerical value, and the reciprocal (I) is obtainedT·Ω-1·I)-1(ii) a (4) Comparing the result of (3) with ITMultiplying; (5) comparing the result in (4) with omega-1Multiplying; (6) multiplying the result in (5) by y, thereby obtaining a b value;
4. finding alphaiValue can be given by the formula α ═ Ω-1(y-b I) performing a calculation; the method specifically comprises the following steps: (1) b, calculating b I; (2) calculating y-b I; (3) will omega-1Multiplying the result of the above (2) to obtain alphaiA value;
5. further, the above b value and alpha are measurediValue substitution prediction function
Calculating a predicted value; wherein,
i is a natural number from 1 to N.
Assume that the variables are 3, sample 1 (corresponding to matrix x)1) Respectively as follows: a1, a2, a 3; sample 2 (correspondence matrix x)2) Respectively as follows: b1, b2, b 3; then And is <math>
<mrow>
<msubsup>
<mi>x</mi>
<mn>1</mn>
<mi>T</mi>
</msubsup>
<mo>·</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>=</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>a</mi>
<mrow>
<mn>1</mn>
<mo>,</mo>
</mrow>
</msub>
<msub>
<mi>a</mi>
<mn>2</mn>
</msub>
<mo>,</mo>
<msub>
<mi>a</mi>
<mn>3</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>·</mo>
<mfenced open='(' close=')'>
<mtable>
<mtr>
<mtd>
<msub>
<mi>a</mi>
<mn>1</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>a</mi>
<mn>2</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>a</mi>
<mn>3</mn>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
</math> <math>
<mrow>
<msubsup>
<mi>x</mi>
<mn>1</mn>
<mi>T</mi>
</msubsup>
<mo>·</mo>
<msub>
<mi>x</mi>
<mn>2</mn>
</msub>
<mo>=</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>a</mi>
<mrow>
<mn>1</mn>
<mo>,</mo>
</mrow>
</msub>
<msub>
<mi>a</mi>
<mn>2</mn>
</msub>
<mo>,</mo>
<msub>
<mi>a</mi>
<mn>3</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>·</mo>
<mfenced open='(' close=')'>
<mtable>
<mtr>
<mtd>
<msub>
<mi>b</mi>
<mn>1</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>b</mi>
<mn>2</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>b</mi>
<mn>3</mn>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
</math> When there are n samples, and so on.
The system 20 for passenger flow data analysis may be configured to acquire historical passenger flow data to construct a sample set of an LS _ SVM model, perform regression training on the sample set by using an LS _ SVM algorithm, calculate relevant parameters of a regression function to construct a regression function, input preset prediction variable values to the regression function, and calculate corresponding predicted passenger flow data, thereby obtaining more accurate prediction data.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method of passenger flow data analysis, comprising the steps of:
collecting historical passenger flow data containing pre-estimated variables;
establishing a pre-estimation model according to the historical passenger flow data;
and estimating passenger flow data according to the estimated variables and the estimated model.
2. The method of passenger flow data analysis according to claim 1, further comprising:
and setting an estimated variable of the passenger flow data.
3. The method of passenger flow data analysis according to claim 2, wherein the predictor variables include weather conditions and holiday conditions.
4. The method for passenger flow data analysis according to any of claims 1 to 3, wherein the step of estimating the passenger flow data based on the pre-estimated variables and the pre-estimated model comprises:
performing regression training on the historical data by adopting an LS _ SVM algorithm, calculating relevant parameters of a regression function, and constructing the regression function;
and putting the estimated variables into the regression function to obtain the estimated passenger flow data.
5. The method of passenger flow data analysis according to claim 4, wherein the step of calculating regression function related parameters and constructing a regression function specifically comprises:
calculating to obtain corresponding parameters b and alphaiAnd constructing a prediction function:
6. a system for passenger flow data analysis, comprising:
the data acquisition unit is used for acquiring historical passenger flow data containing the pre-estimated variables;
the model establishing unit is used for establishing a pre-estimation model according to the historical passenger flow data;
and the estimation calculation unit is used for estimating the passenger flow data according to the estimation variable and the estimation model.
7. The system of passenger flow data analysis of claim 6, further comprising:
and the variable setting unit is used for setting the estimated variable of the passenger flow data.
8. The system of passenger flow data analysis according to claim 7, wherein the predictor variables include weather conditions and holiday conditions.
9. The system for passenger flow data analysis according to any one of claims 6 to 8, wherein the pre-estimation calculation unit specifically comprises:
the function building module is used for carrying out regression training on the historical data by adopting an LS _ SVM algorithm, calculating relevant parameters of a regression function and building the regression function;
and the data acquisition module is used for putting the pre-estimated variables into the regression function to acquire the estimated passenger flow data.
10. The system for passenger flow data analysis of claim 9, wherein the function building module is specifically configured to:
calculating to obtain corresponding parameters b and alphaiAnd constructing a prediction function:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110348316.6A CN102819768B (en) | 2011-11-07 | 2011-11-07 | The method and system that passenger flow data is analyzed |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110348316.6A CN102819768B (en) | 2011-11-07 | 2011-11-07 | The method and system that passenger flow data is analyzed |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102819768A true CN102819768A (en) | 2012-12-12 |
| CN102819768B CN102819768B (en) | 2015-08-19 |
Family
ID=47303876
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110348316.6A Active CN102819768B (en) | 2011-11-07 | 2011-11-07 | The method and system that passenger flow data is analyzed |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102819768B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632212A (en) * | 2013-12-11 | 2014-03-12 | 北京交通大学 | System and method for predicating time-varying user dynamic equilibrium network-evolved passenger flow |
| CN103793761A (en) * | 2014-01-24 | 2014-05-14 | 同济大学 | Method for identifying hub abnormal passenger flow volume generating factors |
| CN107169555A (en) * | 2017-06-30 | 2017-09-15 | 广东欧珀移动通信有限公司 | A kind of gate reminding method, device, storage medium and terminal |
| CN107180270A (en) * | 2016-03-12 | 2017-09-19 | 上海宏理信息科技有限公司 | Passenger flow forecasting and system |
| CN107180278A (en) * | 2017-05-27 | 2017-09-19 | 重庆大学 | A kind of real-time passenger flow forecasting of track traffic |
| CN108376292A (en) * | 2017-12-12 | 2018-08-07 | 广州汇智通信技术有限公司 | A kind of crowd's method for predicting, system and equipment |
| CN108596401A (en) * | 2016-11-25 | 2018-09-28 | 口碑(上海)信息技术有限公司 | A kind of prediction technique and device of portfolio |
| CN109190546A (en) * | 2018-08-28 | 2019-01-11 | 广州洪荒智能科技有限公司 | One kind being based on computer vision bus station stream of people's analysis method |
| CN109308543A (en) * | 2018-08-20 | 2019-02-05 | 华南理工大学 | The short-term passenger flow forecasting of subway based on LS-SVM and real-time big data |
| CN109559512A (en) * | 2018-12-05 | 2019-04-02 | 北京掌行通信息技术有限公司 | A kind of regional traffic flow prediction technique and device |
| CN111210047A (en) * | 2019-11-12 | 2020-05-29 | 恒大智慧科技有限公司 | Scenic spot service time estimation method, device and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7251589B1 (en) * | 2005-05-09 | 2007-07-31 | Sas Institute Inc. | Computer-implemented system and method for generating forecasts |
| CN102034350A (en) * | 2009-09-30 | 2011-04-27 | 北京四通智能交通系统集成有限公司 | Short-time prediction method and system of traffic flow data |
-
2011
- 2011-11-07 CN CN201110348316.6A patent/CN102819768B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7251589B1 (en) * | 2005-05-09 | 2007-07-31 | Sas Institute Inc. | Computer-implemented system and method for generating forecasts |
| CN102034350A (en) * | 2009-09-30 | 2011-04-27 | 北京四通智能交通系统集成有限公司 | Short-time prediction method and system of traffic flow data |
Non-Patent Citations (1)
| Title |
|---|
| 朱家元等: "基于优化最小二乘支持向量机的小样本预测研究", 《航空学报》, vol. 25, no. 6, 30 November 2004 (2004-11-30), pages 565 - 568 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632212B (en) * | 2013-12-11 | 2017-01-25 | 北京交通大学 | System and method for predicating time-varying user dynamic equilibrium network-evolved passenger flow |
| CN103632212A (en) * | 2013-12-11 | 2014-03-12 | 北京交通大学 | System and method for predicating time-varying user dynamic equilibrium network-evolved passenger flow |
| CN103793761A (en) * | 2014-01-24 | 2014-05-14 | 同济大学 | Method for identifying hub abnormal passenger flow volume generating factors |
| CN107180270A (en) * | 2016-03-12 | 2017-09-19 | 上海宏理信息科技有限公司 | Passenger flow forecasting and system |
| CN108921333A (en) * | 2016-11-25 | 2018-11-30 | 口碑(上海)信息技术有限公司 | A kind of prediction technique and device of portfolio |
| CN108596401A (en) * | 2016-11-25 | 2018-09-28 | 口碑(上海)信息技术有限公司 | A kind of prediction technique and device of portfolio |
| CN108596401B (en) * | 2016-11-25 | 2020-04-24 | 口碑(上海)信息技术有限公司 | Traffic prediction method and device |
| CN108921333B (en) * | 2016-11-25 | 2021-04-27 | 口碑(上海)信息技术有限公司 | Traffic prediction method and device |
| US11443251B2 (en) | 2016-11-25 | 2022-09-13 | Koubei Holding Limited | Method and device for predicting traffic |
| CN107180278A (en) * | 2017-05-27 | 2017-09-19 | 重庆大学 | A kind of real-time passenger flow forecasting of track traffic |
| CN107169555A (en) * | 2017-06-30 | 2017-09-15 | 广东欧珀移动通信有限公司 | A kind of gate reminding method, device, storage medium and terminal |
| CN108376292A (en) * | 2017-12-12 | 2018-08-07 | 广州汇智通信技术有限公司 | A kind of crowd's method for predicting, system and equipment |
| CN109308543A (en) * | 2018-08-20 | 2019-02-05 | 华南理工大学 | The short-term passenger flow forecasting of subway based on LS-SVM and real-time big data |
| CN109308543B (en) * | 2018-08-20 | 2021-10-22 | 华南理工大学 | Subway short-term passenger flow prediction method based on LS-SVM and real-time big data |
| CN109190546A (en) * | 2018-08-28 | 2019-01-11 | 广州洪荒智能科技有限公司 | One kind being based on computer vision bus station stream of people's analysis method |
| CN109559512A (en) * | 2018-12-05 | 2019-04-02 | 北京掌行通信息技术有限公司 | A kind of regional traffic flow prediction technique and device |
| CN111210047A (en) * | 2019-11-12 | 2020-05-29 | 恒大智慧科技有限公司 | Scenic spot service time estimation method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102819768B (en) | 2015-08-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102819768B (en) | The method and system that passenger flow data is analyzed | |
| CN107193876B (en) | Missing data filling method based on nearest neighbor KNN algorithm | |
| Tsui et al. | Estimating airport efficiency of New Zealand airports | |
| Cuadrado-Gallego et al. | An experimental study on the conversion between IFPUG and COSMIC functional size measurement units | |
| EP3138075A2 (en) | Decision support system for hospital quality assessment | |
| CN104484582B (en) | The biological information project automatic analysis method and system realized by modularization selection | |
| CN113807747A (en) | Enterprise budget management maturity evaluation system | |
| CN105427194A (en) | Method and device for electricity sales amount prediction based on random forest regression | |
| WO2017071369A1 (en) | Method and device for predicting user unsubscription | |
| CN110111036A (en) | Logistics goods amount prediction technique and system based on LSTM Model Fusion | |
| Valeva et al. | A matheuristic for workforce planning with employee learning and stochastic demand | |
| CN104008433A (en) | Method for predicting medium-and-long-term power loads on basis of Bayes dynamic model | |
| CN107563560A (en) | A kind of Traffic prediction method, apparatus, equipment, storage medium | |
| CN111415027A (en) | Method and device for constructing component prediction model | |
| US20170372331A1 (en) | Marking of business district information of a merchant | |
| Martin et al. | Using event logs to model interarrival times in business process simulation | |
| Soorapanth et al. | Towards a framework for evaluating the costs and benefits of simulation modelling in healthcare | |
| CN111353797B (en) | Resource allocation method and device and electronic equipment | |
| CN110852322B (en) | Method and device for determining region of interest | |
| Naumann et al. | Automation in high‐content flow cytometry screening | |
| RU2586025C2 (en) | Method for automatic clustering of objects | |
| CN108880835A (en) | Data analysing method and device, computer storage medium | |
| CN113779391A (en) | Intelligent lock unlocking recommendation method, system and device based on modeling and storage medium | |
| CN110689158B (en) | Method, device and storage medium for predicting destination | |
| CN112308344A (en) | Method and device for predicting reservation value of non-departure flight and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |