WO2015158149A1 - 一种用于对特征信息的变化进行预测的方法和装置 - Google Patents
一种用于对特征信息的变化进行预测的方法和装置 Download PDFInfo
- Publication number
- WO2015158149A1 WO2015158149A1 PCT/CN2014/093952 CN2014093952W WO2015158149A1 WO 2015158149 A1 WO2015158149 A1 WO 2015158149A1 CN 2014093952 W CN2014093952 W CN 2014093952W WO 2015158149 A1 WO2015158149 A1 WO 2015158149A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- feature
- feature information
- change
- user
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Definitions
- step S1 the estimating device acquires historical feature data and current incremental data of the plurality of first feature information corresponding to the estimated model in at least one calculation period.
- the prediction model includes, but is not limited to, a model for predicting a change of the second feature information based on the at least one first feature information.
- the estimating device may first obtain a plurality of periodic average values of the feature data of the first feature information in each calculation period, and further calculate based on the obtained plurality of over-period average values.
- the second average of the period averages is taken as the historical feature data of the first feature information with respect to the plurality of calculation periods.
- the person skilled in the art can select or determine the length of the calculation period based on actual conditions and requirements, for example, 7 days as one calculation period, and, for example, one month as one calculation period or the like.
- step S5 the estimating means acquires the feature value of the second feature information every day between the current starting point to the historical time period.
- the second feature information is a consumption value of the user
- the predetermined history time period has a length of 7 days
- the prediction date is the xth day
- the current day is Day x-1.
- the first feature information for establishing the estimated model model_1 includes: a bid of the keyword, a click amount of the advertisement corresponding to the keyword, a number of periods of the advertisement corresponding to the keyword, and a number of regions. Then, the estimating device acquires, in step S4, the average value of each of the first feature information "bid", “click amount”, "number of time slots” and “number of regions” during the period from day x-2 to day x-8, respectively.
- Each of the first feature information respectively corresponds to the increment information
- the estimating means acquires the feature value of the second feature information "consumption value" every day from the xth day to the x-8th day in step S5. Then, the estimating device according to the obtained first feature information on the x-th day to the x-th day of the historical feature data and the corresponding incremental information and the second feature information consumption value on the x-1th day
- the eigenvalue S x-1 gives the estimated model model_1 for predicting the consumption value on the xth day.
- the estimating device performs the above steps S4 and steps.
- S5 an estimated model model_2 for predicting the amount of clicks on the xth day is obtained.
- any one or more first feature information is in the historical time period.
- the historical feature data and the corresponding incremental information therein, and the obtained feature value of the second feature information per day, determining an estimated model for predicting the feature value of the second feature information on the next day Implementations are intended to be included within the scope of the present invention.
- step S2 the estimation device acquires the prediction model based on the historical feature data of the first feature information corresponding thereto and the current incremental data for prediction processing.
- the first change information of the second feature information determined on the predicted date.
- step S3 the estimating device determines, according to the first change information, change estimation information of the second feature information on the predicted date, to prompt the user to perform corresponding according to the change prediction information. operating.
- the change prediction information further includes other information for indicating a change of the final estimated value of the second feature information with respect to the previous feature value, for example, for indicating the second feature information.
- the estimating device first determines second change information of the second feature information on the predicted date according to the historical feature value of the second feature information; and then, the estimating device is configured according to the first change information and the first The second change information determines the change estimate information of the second feature information on the predicted date to prompt the user to perform a corresponding operation based on the change estimate information.
- the second consumption information is used to indicate an estimated value of the second feature information determined according to the historical feature value of the second feature information.
- the estimating means may determine the second change information of the second feature information on the predicted date based on the following formula (1):
- sn represents second variation information of the second feature information on the nth day
- s1 represents the feature value of the second feature information on the first day after the current time
- s1 is equal to the average value of the historical feature values of the past 7 days
- wn Indicates the number of days in the period to which the nth day belongs (for example, 7 days is a calculation period, then the number of days in the first day of the period is 1, and the number of days in the second day is 2,.. and so on, The number of days in the seven-day cycle is 7)
- swn The average value of the historical eigenvalues of all days in the past, in which the number of days in the cycle is w1.
- the estimating device processes the first change information and the second change information according to a predetermined processing rule to obtain the second feature information on the predicted date. Change estimate information.
- the estimating device further acquires the weights of the first change information and the second change information respectively, so that the weighted sum of the first change information and the second change information is used as the change estimate information of the second feature information on the predicted date. And prompting the user to perform a corresponding operation based on the change estimation information.
- the estimating device may use the change estimate information of the predicted date as a historical feature value of the second feature information, and the predicted date is One day is used as a new prediction date, and the change estimation information of the second feature information on the new predicted date is determined by repeatedly performing steps S1 to S3. Similarly, the estimating device may repeatedly perform steps S1 to S3 in this manner to determine change estimation information for a plurality of days in the future.
- the method further comprises a step S7 (not shown).
- step S7 when an update operation of at least one of the plurality of first feature information corresponding to the estimated model by the user is obtained, the estimating device updates the according to the update operation.
- the current incremental data of each of the at least one first feature information is obtained to obtain an updated prediction model.
- the update operation includes various operations for changing a feature value of the first feature information.
- the estimating device acquires that the value of the first feature information "bid” is changed from price_0 to price_1 by the user, and the estimating device is based on the "bid” in the past x-1 day to the first
- the average price_e of the first feature information "bid” is updated from price_0/price_e to price_1/price_e for the estimated model model_1 based on the current status of the updated "bid”. Incremental data to perform prediction processing.
- the method according to the invention further comprises a step S8 (not shown), the step S3 further comprising a step S301 (not shown).
- step S8 the estimating device acquires an estimated presentation instruction of the current user.
- the estimated presentation instruction includes but is not limited to at least one of the following:
- change prediction information of one or more second feature information is selected among the plurality of second feature information for presentation.
- the estimating device displays two predictable two second feature information “consumption value” and “retrieve amount” on the user interface, and acquires change estimation information that the user selects to present the first feature information “consumption value”.
- a presentation form of the change prediction information for example, presentation in the form of a graph, or presentation in the form of a data list;
- the time period of the change prediction information presented for example, the next week or the next month, etc., preferably, the user can select any one or more days in the future as the time period for presenting the change estimate information.
- step S301 the estimating device presents according to the estimated rendering instruction. At least one change estimate information corresponding to the estimated presentation instruction.
- the estimating device acquiring the estimated presentation instruction of the current user includes: presenting the feature value of the consumption value in the past 10 days and the change prediction information of the next 10 days; and presenting in the form of a graph. Then, the estimating device acquires information corresponding to the estimated rendering instruction from the historical feature values of the plurality of second feature information that have been obtained, and performs the foregoing steps S1 to S3 multiple times to obtain the second feature information.
- the estimated value of the "consumption value" in the next 10 days is presented in the form of a graph to show the user the graph shown in FIG.
- the method according to the present embodiment further includes step S9 (not shown), step S10 (not shown), and step S11 (not shown).
- step S9 the estimation device estimates the change prediction information of the at least one second feature information of the plurality of users in the predetermined time period.
- step S10 the estimating device matches the respective change estimation information of the plurality of users with the change estimation information of the current user within the predetermined time period to obtain according to the fitting result. Determining one or more similar users of the current user.
- the estimating device determines a feature update plan for recommending to the current user according to the feature update plan corresponding to the one or more similar users and the second feature information. For example, the estimating device fits a curve of the estimated change information of the “consumption value” of the plurality of users obtained in the future period of time to obtain a user that best matches the current user, and acquires the user in the future. A consumption plan within five days (ie, a feature update plan corresponding to the first feature information) to recommend the consumption plan to the current user.
- the method according to the invention further comprises a step S12 (not shown).
- step S12 the estimating device determines whether the change prediction information satisfies a predetermined prompt condition, and when the predetermined prompt condition is met, sends a corresponding prompt information to the current user to prompt the user to perform a corresponding operation.
- the change estimate information includes an estimated feature value of the user's consumption value
- the predetermined prompt condition includes: the account balance of the forecast date is less than 1000 yuan.
- the current account balance of the user is 1500 yuan
- the estimating device obtains the estimated feature value of the consumption value on the forecast date by 820 yuan by performing the foregoing steps S1 to S3, and the estimating device determines the user according to the change estimated value.
- the account balance on the forecast date is 680 yuan.
- the estimating device determines that the account balance corresponding to the estimated value of the change satisfies the predetermined prompt condition, and sends a prompt message of “the balance is insufficient, please recharge in time” to the current user, so as to prompt the current user to perform the operation of recharging the account.
- the method according to the invention further comprises a step S13 (not shown) and a step S14 (not shown).
- the estimating apparatus performs at least one of the foregoing steps S1 to S3 on the one or more users to obtain a change pre-preparation of the at least one second feature information respectively corresponding to each of the plurality of users. And/or the estimating device receives the change estimate information of the at least one second feature information respectively corresponding to the one or more users obtained by the other estimating device.
- step S14 the estimating device counts the obtained change prediction information of the at least one second feature information corresponding to each user to adjust the corresponding service resource configuration based on the statistical result.
- the resource includes various types of hardware and software resources required for providing a service related to the feature information.
- the service resources include the available bandwidth size, and for example, include the number of available servers, and the like.
- the second feature information includes a “search amount”, and in step S13, the estimating device receives the search amount of each user corresponding to the plurality of other estimating devices from a plurality of other estimating devices within a next week period.
- An alternate server for the service to increase the throughput of the retrieval service during this period.
- the method of the present invention by constructing an estimation model for predicting the second feature information based on the plurality of first feature information, it is possible to effectively reflect the mutual influence between the respective feature information. Relationship, so that the corresponding feature information can be estimated according to the estimated model in the future, so that the user can understand the future change trend of the feature information, and perform corresponding operations based on the presented estimated information, thereby improving the user experience; By combining the operations performed by the user to estimate the feature value of the second feature information in a future period of time, the accuracy of the estimation information is further improved, and the user's desire to understand the future change of the feature information is satisfied, and the requirement is improved. Estimated accuracy.
- Fig. 2 is a block diagram showing the construction of an estimating apparatus for predicting changes in feature information according to the present invention.
- the estimating device includes: means for acquiring historical feature data and current incremental data of the plurality of first feature information corresponding to the estimated model in at least one calculation period (hereinafter referred to as "first acquiring device 1" ⁇ / RTI>; for obtaining the prediction model based on the historical feature data of the first feature information corresponding thereto and the current incremental data, the second feature information is determined on the prediction date a device for changing information (hereinafter referred to as "second acquisition device 2"); for determining change estimation information of the second feature information on the prediction date according to the first change information, based on the change
- the device that estimates the information to prompt the user to perform the corresponding operation hereinafter referred to as "determination device 3").
- the first obtaining device 1 acquires historical feature data and current incremental data of the plurality of first feature information corresponding to the estimated model in at least one calculation period.
- the prediction model includes, but is not limited to, a model for predicting a change of the second feature information based on the at least one first feature information.
- the predictive model can be implemented in a machine learning manner.
- the manner in which the first acquiring device 1 acquires the first feature information and the second feature information includes, but is not limited to, any one of the following:
- one of the predetermined feature information is predetermined to be predicted, and the remaining feature information is the first feature information.
- the predetermined “consumption value” is the second feature information to be predicted, and the remaining feature information "retrieve amount” and "rank” are required for establishing an estimation model for predicting the second feature information.
- First feature information is the second feature information to be predicted, and the remaining feature information “retrieve amount” and “rank” are required for establishing an estimation model for predicting the second feature information.
- the first acquisition device 1 selects one of the plurality of feature information as the second feature information according to the user operation, and selects at least one of the remaining feature information as the first feature information.
- the historical feature data includes a periodic average value of the feature values of the first feature information in each calculation period.
- the first obtaining device 1 may first obtain a plurality of periodic average values of the feature data of the first feature information in each calculation period, and then based on the obtained plurality of over-period average values, A quadratic average of the plurality of period averages is calculated and used as historical feature data of the first feature information with respect to the plurality of calculation periods.
- the person skilled in the art can select or determine the length of the calculation period based on actual conditions and requirements, for example, 7 days as one calculation period, and, for example, one month as one calculation period or the like.
- the current incremental data is used to indicate a ratio of the first feature data of the first feature information on a day before the predicted date to the historical feature data of the at least one calculation cycle.
- the first acquiring device 1 acquires feature values of each of the plurality of first feature information corresponding to the predicted model in each of the at least one calculation period to determine each of the corresponding models The historical feature data of the first feature information in the at least one calculation period and the respective current incremental data.
- the estimating device may determine the predictive model by acquiring historical feature data of one or more first feature information in a historical time period and corresponding increments thereof respectively Means for information (not shown, hereinafter referred to as "first sub-acquisition device"), means for estimating the device to acquire the characteristic value of the second feature information between the current and the starting point of the historical time period (not shown) Shown below, referred to as “second sub-acquisition device”) and historical feature data for historical time period according to the one or more first feature information and their corresponding incremental information, and the obtained obtained obtained
- the feature value of the second feature information per day is determined, and means for predicting the prediction model of the feature value of the second feature information on the next day (not shown, referred to as "sub-determination device” hereinafter).
- the first sub-acquisition device acquires historical feature data of the one or more first feature information in the historical time period and the corresponding incremental information.
- the historical time period includes at least one calculation period.
- the second sub-acquisition device acquires the feature value of the second feature information every day between the current and the starting point of the historical time period.
- the sub-determining device according to the historical feature data of the one or more first feature information in the historical time period and the corresponding incremental information, and the obtained feature value of the second feature information per day, An estimation model for predicting the feature value of the second feature information on the next day is determined.
- the estimating device may: according to the historical feature data of the one or more first feature information in the historical time period and the corresponding incremental information, and the acquired feature value of the second feature information,
- the regression model is used to determine the prediction model. For example, vector regression, random forest, linear regression and other regression analysis.
- the second feature information is a consumption value of the user
- the predetermined history time period has a length of 7 days
- the prediction date is the xth day
- the current day is Day x-1.
- the first feature information for establishing the estimated model model_1 includes: a bid of the keyword, a click amount of the advertisement corresponding to the keyword, a number of periods of the advertisement corresponding to the keyword, and a number of regions.
- the first sub-acquisition device acquires an average value of each of the first feature information "bid”, “click amount”, “number of time slots”, and “number of regions” during the period from the x-2th day to the x-8th day, and each The feature information respectively corresponds to the increment information, and the second sub-acquisition device acquires the feature value of the second feature information "consumption value" every day from the xth day to the x-8th day.
- the sub-determining device according to the obtained first feature information on the x-th day to the x-th day of the historical feature data and the corresponding incremental information and the second feature information consumption value on the xth day
- the eigenvalue S x-1 gives the estimated model model_1 for predicting the consumption value on the xth day.
- the estimating device performs the above steps S4 and steps.
- S5 an estimated model model_2 for predicting the amount of clicks on the xth day is obtained.
- any one or more first feature information is in the historical time period.
- the historical feature data and the corresponding incremental information therein, and the obtained feature value of the second feature information per day, determining an estimated model for predicting the feature value of the second feature information on the next day Implementations are intended to be included within the scope of the present invention.
- the second obtaining means 2 acquires the second feature information determined by the prediction model based on the historical feature data of the first feature information corresponding to the first feature information and the current incremental data, and the second feature information is on the predicted date.
- the first change information is used to determine the second feature information.
- the first change information includes, but is not limited to, an estimated feature value of the second feature information to be predicted on the predicted date.
- the first change information further includes other information for indicating a change of the estimated feature value of the second feature information with respect to the previous feature value, for example, for indicating the second feature information.
- the calculation cycle length is 7 days and the prediction date is the xth day.
- the first obtaining device 1 acquires the first feature information “bid”, “click amount”, “number of time slots” and “number of regions” corresponding to the current user according to the determined prediction model model_1, respectively, on the x-1th day to the first Mean values in x-7 days: price_e, click_e, time_e, and area_e, and the first obtaining means 1 determines each first based on the feature values of the respective first feature information on the x-1th day: price_0, click_0, time_0, and area_0.
- the current incremental data of the feature information price_0/price_e, click_0/click_e, time_0/time_e, and area_0/area_e.
- the second obtaining means 2 obtains the estimated consumption value of the second feature information "consumption value" obtained by the prediction model model_1 based on the average value of the respective first feature information and the current incremental data after the prediction processing on the xth day. S x .
- the first obtaining means 1 acquires the first feature information "bid”, “consumption value”, “number of time slots” and “number of regions” according to the determined prediction model model_2, respectively, on the xth day to the x-th Averages over 7 days: price_e, cost_e, time_e, and area_e, and their respective current delta data: price_0/price_e, cost_0/cost_e, time_0/time_e, and area_0/area_e.
- the second obtaining means 2 obtains the estimated hit amount C of the second feature information "click amount" obtained by the estimated model model_2 based on the average value of each of the first feature information and the current incremental data. x .
- step S3 the determining device 3 determines, according to the first change information, change estimation information of the second feature information on the predicted date, to prompt the user to perform corresponding according to the change prediction information. operating.
- the change prediction information includes, but is not limited to, a final estimated value of the second feature information.
- the change prediction information further includes other information for indicating a change of the final estimated value of the second feature information with respect to the previous feature value, for example, for indicating the second feature information.
- the determining apparatus 3 determines, according to the first change information, change estimation information of the second feature information on the predicted date, to prompt the user to perform a corresponding operation based on the change prediction information.
- the method includes any of the following:
- the determining device 3 uses the obtained estimated consumption value S x and the estimated click amount C x as the change prediction information of the xth day to prompt the user to perform the corresponding based on the change estimation information. Operation. For example, prompt the user to recharge in time, or recommend a consumption plan suitable for him to the current user.
- the estimating device further includes: means for determining second change information of the second feature information on the predicted date according to the historical feature value of the second feature information (not shown, referred to as "the first The third obtaining means "); the determining means 3 further comprising: determining, according to the first change information and the second change information, change estimate information of the second feature information on a predicted date to be based on the change
- the device for estimating the information to prompt the user to perform the corresponding operation (not shown, hereinafter referred to as "sub-determination device").
- the second consumption information is used to indicate an estimated value of the second feature information determined according to the historical feature value of the second feature information.
- the third obtaining means averages the historical feature values over a period of time in the past The value is used as the second change information of the forecast day.
- the third obtaining means may determine the second change information of the second feature information on the predicted date based on the following formula (1):
- sn represents second variation information of the second feature information on the nth day
- s1 represents the feature value of the second feature information on the first day after the current time
- s1 is equal to the average value of the historical feature values of the past 7 days
- wn Indicates the number of days in the period to which the nth day belongs (for example, 7 days is a calculation period, then the number of days in the first day of the period is 1, and the number of days in the second day is 2,.. and so on, The number of days in the seven-day cycle is 7)
- swn represents the average of the historical eigenvalues of all days in the past in the plurality of calculation cycles.
- the sub-determination device processes the first change information and the second change information according to a predetermined processing rule to obtain the second feature information on the predicted date. Change estimate information.
- the sub-determining device uses the sum of the first change information and the second change information of the second feature information as the change estimate information of the second feature information on the predicted date to prompt the user to perform based on the change estimate information.
- the corresponding operation uses the sum of the first change information and the second change information of the second feature information as the change estimate information of the second feature information on the predicted date to prompt the user to perform based on the change estimate information.
- the sub-determining device further acquires the weights of the first change information and the second change information, respectively, to use the weighted sum of the first change information and the second change information as the change estimate information of the second feature information on the predicted date. And prompting the user to perform a corresponding operation based on the change estimation information.
- the determining device 3 may use the change estimate information of the predicted date as a historical feature value of the second feature information, and the predicted date is One day as a new forecast day, and through repeated execution And taking the operation of the historical feature data and the current incremental data of the plurality of first feature information corresponding to the prediction model in the at least one calculation period to determine, according to the first change information, the second feature information in the prediction The change of the day predicts the operation of the information to determine the change estimate information of the second feature information on the new predicted date.
- the estimating device may repeatedly perform the operation of acquiring the historical feature data and the current incremental data of the plurality of first feature information corresponding to the estimated model in the at least one calculation period to the first according to the first
- the change information determines an operation of the change estimate information of the second feature information on the predicted date to determine change estimate information for a plurality of days in the future.
- the estimating means further includes an update operation for obtaining at least one of the plurality of pieces of first feature information corresponding to the estimated model by the user And updating, by the update operation, the current incremental data of each of the at least one first feature information to obtain an updated prediction model (not shown, referred to as “feature update device” hereinafter).
- the feature update device updates the at least one first according to the update operation
- the current incremental data of the feature information is obtained to obtain an updated prediction model.
- the update operation includes various operations for changing a feature value of the first feature information.
- the estimating device acquires that the value of the first feature information "bid” is changed from price_0 to price_1 by the user, and the feature updating device is based on the "bid” in the past x-1 day to the first
- the average price_e of the first feature information "bid” is updated from price_0/price_e to price_1/price_e for the estimated model model_1 based on the current status of the updated "bid”. Incremental data to perform prediction processing.
- the estimating apparatus further comprises means for acquiring an estimated presentation instruction of the current user (not shown, hereinafter referred to as “instruction acquiring means"), the determining means 3 further comprising And presenting an instruction to present at least one piece of change estimate information corresponding to the estimated presence instruction (not shown, referred to as “presentation device” hereinafter).
- instruction acquiring means means for acquiring an estimated presentation instruction of the current user
- determining means 3 further comprising And presenting an instruction to present at least one piece of change estimate information corresponding to the estimated presence instruction (not shown, referred to as “presentation device” hereinafter).
- the instruction acquisition device acquires an estimated presentation instruction of the current user.
- the estimated presentation instruction includes but is not limited to at least one of the following:
- change prediction information of one or more second feature information is selected among the plurality of second feature information for presentation.
- the estimating device displays two predictable two first feature information “consumption value” and “retrieve amount” on the user interface, and acquires change estimation information that the user selects to present the first feature information “consumption value”.
- a presentation form of the change prediction information for example, presentation in the form of a graph, or presentation in the form of a data list;
- the time period of the change prediction information presented for example, the next week or the next month, etc., preferably, the user can select any one or more days in the future as the time period for presenting the change estimate information.
- the rendering device presents at least one piece of change estimation information corresponding to the estimated presentation instruction according to the estimated presentation instruction.
- the instruction acquiring device acquiring the estimated presentation instruction of the current user includes: presenting the feature value of the consumption value in the past 10 days and the change prediction information of the next 10 days; and presenting in the form of a graph.
- the rendering device acquires information corresponding to the estimated presentation instruction from the historical feature values of the obtained second feature information, and performs the foregoing acquiring the plurality of second feature information corresponding to the prediction model by using the foregoing multiple times.
- the characteristic information "consumption value" is estimated information of the change in the next 10 days, and is presented in the form of a graph to present the graph shown in FIG. 3 to the user.
- the estimating apparatus further includes: means for respectively estimating change information of at least one second feature information of the plurality of users within the predetermined time period (not shown) Shown below, referred to as "predetermined device”; for fitting the respective change estimation information of the plurality of users with the change estimation information of the current user within the predetermined time period, according to the a device for fitting one or more similar users of the current user (not shown, referred to as “fitting device” for short); and for determining one or more similar users and the second feature according to the one or more similar users
- the feature update plan corresponding to the information determines a feature update plan for recommending to the current user.
- the device (not shown, hereinafter referred to as "plan recommendation device”).
- the predetermined estimating information of the at least one second feature information of the plurality of users in the predetermined time period by the predetermined device is not limited.
- the fitting device fits the respective change estimation information of the plurality of users with the change estimation information of the current user within the predetermined time period to determine the according to the fitting result.
- One or more similar users of the current user One or more similar users of the current user.
- the plan recommendation device determines a feature update plan for recommending to the current user according to the feature update plan corresponding to the one or more similar users and the second feature information.
- the fitting device fits the obtained curve of the change estimate information of the “consumption value” of the plurality of users in the future period of time to obtain a user that best matches the current user, and acquires the user in the future.
- a consumption plan within five days ie, a feature update plan corresponding to the first feature information
- the plan recommendation device to recommend the consumption plan to the current user.
- the estimating apparatus further comprises: determining whether the change prediction information satisfies a predetermined prompt condition, and when the predetermined prompt condition is met, transmitting corresponding prompt information to the current user to prompt the user to perform the corresponding The device to be operated (not shown, hereinafter referred to as "cue device").
- the prompting device determines whether the change prediction information satisfies a predetermined prompt condition. When the predetermined prompt condition is met, the prompt information is sent to the current user to prompt the user to perform a corresponding operation.
- the change estimate information includes an estimated feature value of the user's consumption value
- the predetermined prompt condition includes: the account balance of the forecast date is less than 1000 yuan.
- the current account balance of the user is 1500 yuan
- the estimating device performs the operation of the historical feature data and the current incremental data in the at least one calculation period by performing the foregoing acquiring the plurality of first feature information corresponding to the prediction model to the Determining, by the change information, the estimated feature value of the operation value obtained by the operation of the change information of the second feature information on the predicted date is 820 yuan, and the estimating device determines the The user's account balance on the forecast date is 680 yuan.
- the prompting device determines that the account balance corresponding to the estimated value of the change satisfies the predetermined prompt condition, and sends a prompt message of “the balance is insufficient, please recharge in time” to the current user, to prompt the current user to perform the operation of recharging the account.
- the estimating apparatus further comprises: means for acquiring change estimation information of at least one piece of second feature information respectively corresponding to one or more users (not shown, hereinafter referred to as "estimated acquisition” And means for adjusting the change estimate information of the at least one second feature information corresponding to each obtained user to adjust the corresponding service resource configuration based on the statistical result (not shown, referred to as “pre Estimate the statistical device”).
- the estimation acquiring device acquires the change estimation information of the at least one second feature information corresponding to the one or more users respectively.
- the estimation acquiring device performs the operation of the historical feature data and the current incremental data in the at least one calculation period by using the plurality of first feature information corresponding to the foregoing acquisition prediction model to the one or more users at least once. Up to determining, according to the first change information, an operation of the change estimate information of the second feature information on the predicted date to obtain a change pre-preparation of at least one second feature information respectively corresponding to each of the plurality of users And the estimated acquisition means receives the change estimation information of the at least one second characteristic information respectively corresponding to the one or more users obtained by the other estimating means.
- the estimation device estimates the change estimation information of the at least one second feature information corresponding to each user obtained by the statistical device to adjust the corresponding service resource configuration based on the statistical result.
- the resource includes various types of hardware and software resources required for providing a service related to the feature information.
- the service resource includes an available bandwidth size, and for example, Includes the number of available servers and more.
- the second feature information includes a “search volume”, and the estimated acquisition device receives the change estimate of the retrieval amount of each user corresponding to the plurality of other estimating devices from the plurality of other estimating devices in the next week. Information, and the change estimate information of the search amount corresponding to the current user obtained by performing the corresponding step in the next week period; then, the estimated statistical device obtains the change estimate information of the search amount corresponding to each user obtained by the statistical device To obtain the sum of the daily average retrieval amounts of the respective users in the next week, and when the sum of the daily average retrieval amounts of the next week exceeds a predetermined threshold, set the standby for providing the retrieval service during the next week. The server to improve the throughput of the retrieval service during this period.
- the mutual influence relationship between the feature information can be effectively embodied, so that the future segment can be predicted according to the prediction model.
- Estimating the corresponding feature information in time facilitating the user to understand the future trend of the feature information, and performing corresponding operations based on the presented estimated information, thereby improving the user experience; and, by combining the operations performed by the user,
- the feature value of the second feature information is estimated in time, which further improves the accuracy of the estimated information, satisfies the user's desire to understand the future change of the feature information, and improves the accuracy of the estimation.
- the corresponding resource configuration can also be adjusted based on the prediction result, so that the global service resources can be utilized more effectively, so that the service corresponding to each feature information can be better supported.
- the software program of the present invention can be executed by a processor to implement the steps or functions described above.
- the software program (including related data structures) of the present invention can be stored in a computer readable recording medium such as a RAM memory, a magnetic or optical drive or a floppy disk and the like.
- some of the steps or functions of the present invention may be implemented in hardware, for example, as a circuit that cooperates with a processor to perform various functions or steps.
- an embodiment in accordance with the present invention includes a device including a memory for storing computer program instructions and a processor for executing program instructions, wherein when the computer program instructions are executed by the processor, triggering
- the apparatus operates based on the aforementioned methods and/or technical solutions in accordance with various embodiments of the present invention.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Evolutionary Computation (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Algebra (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Architecture (AREA)
Abstract
Description
Claims (21)
- 一种用于对特征信息的变化进行预测的方法,其中,所述方法包括以下步骤:-获取预估模型所对应的多项第一特征信息在至少一个计算周期内的历史特征数据及当前增量数据,其中,所述当前增量数据用于指示各个第一特征信息在预测日之前的一天的特征数据相对于所述至少一个计算周期内的历史特征数据之比;-获取预估模型基于其所对应的各项第一特征信息的历史特征数据和所述当前增量数据进行预测处理后所确定的、第二特征信息在所述预测日的第一变化信息-根据所述第一变化信息来确定所述第二特征信息在所述预测日的变化预估信息,以基于所述变化预估信息来提示用户执行相应的操作。
- 根据权利要求1所述的方法,其中,所述方法还包括以下步骤:-根据所述第二特征信息的历史特征值来确定所述第二特征信息在预测日的第二变化信息;其中,所述根据所述第一变化信息来确定所述第二特征信息在所述预测日的变化预估信息,以基于所述变化预估信息来提示用户执行相应的操作的步骤进一步包括以下步骤:-根据所述第一变化信息和所述第二变化信息来确定所述第二特征信息在预测日的变化预估信息,以基于所述变化预估信息来提示用户执行相应的操作。
- 根据权利要求1或2所述的方法,其中,所述获取预估模型所对应的多项第一特征信息的当前增量数据的步骤还包括以下步骤:-当获得用户对所述预估模型对应的所述多项第一特征信息中的至少一项第一特征信息的更新操作时,根据所述更新操作来更新所述至少一项第一特征信息各自的当前增量数据。
- 根据权利要求1至3中任一项所述的方法,其中,所述方法还包 括以下步骤:-获取一项或多项第一特征信息在历史时间段内的历史特征数据及其分别对应的增量信息;其中,所述历史时间段包含至少一个计算周期;-获取第二特征信息在当前至所述历史时间段起始点之间每天的特征值;-根据所述一项或多项第一特征信息在历史时间段内的历史特征数据及其分别对应的增量信息,以及所获得的所述第二特征信息每天的特征值,确定用于预测所述第二特征信息在下一日的特征值的预估模型。
- 根据权利要求1至4中任一项所述的方法,其中,所述方法还包括以下步骤:-根据用户操作,由多个特征信息中选择一个作为第二特征信息,并由余下的特征信息中选择至少一个作为用于建立预测该第二特征信息的预估模型所需的第一特征信息。
- 根据权利要求1至5中任一项所述的方法,其中,所述方法还包括以下步骤:-获取当前用户的预估呈现指令;其中,所述根据所述第一变化信息来确定所述第二特征信息在所述预测日的变化预估信息,以基于所述变化预估信息来提示用户执行相应的操作的步骤还包括以下步骤:-根据所述预估呈现指令,来呈现与所述预估呈现指令对应的至少一项变化预估信息。
- 根据权利要求1至5中任一项所述的方法,其中,所述方法还包括以下步骤:-分别预定时间段内的多个用户的至少一项第二特征信息的变化预估信息;-将所述多个用户的各自的变化预估信息与所述当前用户在所述预定时间段内的变化预估信息进行拟合,以根据所述拟合结果来确定所 述当前用户的一个或多个相似用户;-根据所述一个或多个相似用户的特征更新计划,确定用于向所述当前用户推荐的特征更新计划。
- 根据权利要求1至5中任一项所述的方法,其中,所述方法还包括以下步骤:-判断所述变化预估信息是否满足预定提示条件,当满足预定提示条件时,向所述当前用户发送相应的提示信息以提示用户执行相应的操作。
- 根据权利要求1至8中任一项所述的方法,其中,所述方法还包括以下步骤:-获取与一个或多个用户分别对应的至少一项第二特征信息的变化预估信息;-统计所获得的各个用户对应的至少一项第二特征信息的变化预估信息,以基于统计结果,来调整相应的服务资源配置。
- 一种用于对特征信息的变化进行预测的预估装置,其中,所述预估装置包括:用于获取预估模型所对应的多项第一特征信息在至少一个计算周期内的历史特征数据及当前增量数据的装置,其中,所述当前增量数据用于指示各个第一特征信息在预测日之前的一天的特征数据相对于所述至少一个计算周期内的历史特征数据之比;用于获取预估模型基于其所对应的各项第一特征信息的历史特征数据和所述当前增量数据进行预测处理后所确定的、第二特征信息在所述预测日的第一变化信息的装置;用于根据所述第一变化信息来确定所述第二特征信息在所述预测日的变化预估信息,以基于所述变化预估信息来提示用户执行相应的操作的装置。
- 根据权利要求10所述的预估装置,其中,所述预估装置还包括:用于根据所述第二特征信息的历史特征值来确定所述第二特征信息 在预测日的第二变化信息的装置;其中,所述用于根据所述第一变化信息来确定所述第二特征信息在所述预测日的变化预估信息,以基于所述变化预估信息来提示用户执行相应的操作的装置进一步包括:用于根据所述第一变化信息和所述第二变化信息来确定所述第二特征信息在预测日的变化预估信息,以基于所述变化预估信息来提示用户执行相应的操作的装置。
- 根据权利要求10或11所述的预估装置,其中,所述获取预估模型所对应的多项第一特征信息的当前增量数据的装置还包括:用于当获得用户对所述预估模型对应的所述多项第一特征信息中的至少一项第一特征信息的更新操作时,根据所述更新操作来更新所述至少一项第一特征信息各自的当前增量数据的装置。
- 根据权利要求10至12中任一项所述的预估装置,其中,所述预估装置还包括:用于获取一项或多项第一特征信息在历史时间段内的历史特征数据及其分别对应的增量信息装置;其中,所述历史时间段包含至少一个计算周期;用于获取第二特征信息在当前至所述历史时间段起始点之间每天的特征值的装置;用于根据所述一项或多项第一特征信息在历史时间段内的历史特征数据及其分别对应的增量信息,以及所获得的所述第二特征信息每天的特征值,确定用于预测所述第二特征信息在下一日的特征值的预估模型的装置。
- 根据权利要求10至13中任一项所述的预估装置,其中,所述预估装置还包括:用于根据用户操作,由多个特征信息中选择一个作为第二特征信息,并由余下的特征信息中选择至少一个作为用于建立预测该第二特征信息的预估模型所需的第一特征信息的装置.
- 根据权利要求10至14中任一项所述的预估装置,其中,所述 预估装置还包括:用于获取当前用户的预估呈现指令的装置;其中,所述用于根据所述第一变化信息来确定所述第二特征信息在所述预测日的变化预估信息,以基于所述变化预估信息来提示用户执行相应的操作的装置还包括以:用于根据所述预估呈现指令,来呈现与所述预估呈现指令对应的至少一项变化预估信息的装置。
- 根据权利要求10至14中任一项所述的预估装置,其中,所述预估装置还包括:用于分别预定时间段内的多个用户的至少一项第二特征信息的变化预估信息的装置;用于将所述多个用户的各自的变化预估信息与所述当前用户在所述预定时间段内的变化预估信息进行拟合,以根据所述拟合结果来确定所述当前用户的一个或多个相似用户的装置;用于根据所述一个或多个相似用户的特征更新计划,确定用于向所述当前用户推荐的特征更新计划的装置。
- 根据权利要求10至14中任一项所述的预估装置,其中,所述预估装置还包括:用于判断所述变化预估信息是否满足预定提示条件,当满足预定提示条件时,向所述当前用户发送相应的提示信息以提示用户执行相应的操作装置。
- 根据权利要求10至14中任一项所述的预估装置,其中,所述预估装置还包括:用于获取与一个或多个用户分别对应的至少一项第二特征信息的变化预估信息的装置;用于统计所获得的各个用户对应的至少一项第二特征信息的变化预估信息,以基于统计结果,来调整相应的资源配置的装置。
- 一种计算机可读介质,所述计算机可读介质包括计算机代码,当所述计算机代码被执行时,如权利要求1至9中任一项所述的方法被 执行。
- 一种计算机程序产品,当所述计算机程序产品被计算机设备执行时,如权利要求1至9中任一项所述的方法被执行。
- 一种计算机设备,所述计算机设备包括存储器和处理器,所述存储器中存储有计算机代码,所述处理器被配置来通过执行所述计算机代码以执行如权利要求1至9中任一项所述的方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016550932A JP6254712B2 (ja) | 2014-04-17 | 2014-12-16 | 特徴情報の変化を予測するための方法及び装置 |
EP14889627.7A EP3133537A4 (en) | 2014-04-17 | 2014-12-16 | Method and device for forecasting changes of feature information |
US14/902,303 US10474957B2 (en) | 2014-04-17 | 2014-12-16 | Method and apparatus for forecasting characteristic information change |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410154993.8 | 2014-04-17 | ||
CN201410154993.8A CN103971170B (zh) | 2014-04-17 | 2014-04-17 | 一种用于对特征信息的变化进行预测的方法和装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015158149A1 true WO2015158149A1 (zh) | 2015-10-22 |
Family
ID=51240635
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2014/093952 WO2015158149A1 (zh) | 2014-04-17 | 2014-12-16 | 一种用于对特征信息的变化进行预测的方法和装置 |
Country Status (5)
Country | Link |
---|---|
US (1) | US10474957B2 (zh) |
EP (1) | EP3133537A4 (zh) |
JP (1) | JP6254712B2 (zh) |
CN (1) | CN103971170B (zh) |
WO (1) | WO2015158149A1 (zh) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107507030A (zh) * | 2017-08-22 | 2017-12-22 | 北京京东尚科信息技术有限公司 | 信息预测的方法和装置 |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103971170B (zh) | 2014-04-17 | 2017-09-29 | 北京百度网讯科技有限公司 | 一种用于对特征信息的变化进行预测的方法和装置 |
CN107544981B (zh) * | 2016-06-25 | 2021-06-01 | 华为技术有限公司 | 内容推荐方法及装置 |
CN108229998B (zh) * | 2016-12-21 | 2022-06-03 | 百度在线网络技术(北京)有限公司 | 营销辅助方法及装置 |
CN106875218B (zh) * | 2017-02-04 | 2021-09-28 | 武汉昊阳科技有限公司 | 数据流量产品的价格预测方法及装置 |
CN108197737A (zh) * | 2017-12-29 | 2018-06-22 | 山大地纬软件股份有限公司 | 一种建立医保住院费用预测模型的方法及系统 |
CN108449609B (zh) * | 2018-02-09 | 2020-03-06 | 广州虎牙信息科技有限公司 | 直播间事件的识别方法及装置、电子设备、机器可读介质 |
CN108600970A (zh) * | 2018-03-30 | 2018-09-28 | 深圳春沐源控股有限公司 | 一种信息提醒方法、装置及计算机可读存储介质 |
CN108683734B (zh) * | 2018-05-15 | 2021-04-09 | 广州虎牙信息科技有限公司 | 品类推送方法、装置及存储设备、计算机设备 |
CN109902849B (zh) | 2018-06-20 | 2021-11-30 | 华为技术有限公司 | 用户行为预测方法及装置、行为预测模型训练方法及装置 |
CN110956294B (zh) * | 2018-09-26 | 2021-03-23 | 北京嘀嘀无限科技发展有限公司 | 一种排队时间预估方法以及装置 |
EP3637423A1 (en) * | 2018-10-10 | 2020-04-15 | Koninklijke Philips N.V. | Identifying a user of a display unit |
CN111858015B (zh) * | 2019-04-25 | 2024-01-12 | 中国移动通信集团河北有限公司 | 配置应用程序的运行资源的方法、装置及网关 |
CN112307308A (zh) * | 2019-07-26 | 2021-02-02 | 腾讯科技(深圳)有限公司 | 一种数据处理方法、装置、设备及介质 |
CN111050008A (zh) * | 2019-12-12 | 2020-04-21 | 北京金山云网络技术有限公司 | 一种账户余额提醒方法、装置、电子设备及存储介质 |
CN111062749A (zh) * | 2019-12-12 | 2020-04-24 | 北京爱奇艺科技有限公司 | 增长量预估方法、装置、电子设备及存储介质 |
CN112685360B (zh) * | 2020-12-29 | 2023-09-22 | 湖北华中电力科技开发有限责任公司 | 内存数据的持久化方法及装置、存储介质、计算机设备 |
CN113468235B (zh) * | 2021-05-31 | 2023-05-09 | 北京达佳互联信息技术有限公司 | 信息获取方法、装置、服务器及存储介质 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101673385A (zh) * | 2009-09-28 | 2010-03-17 | 百度在线网络技术(北京)有限公司 | 消费预估方法和装置 |
CN103617459A (zh) * | 2013-12-06 | 2014-03-05 | 李敬泉 | 一种多影响因素下商品需求信息预测方法 |
CN103729351A (zh) * | 2012-10-10 | 2014-04-16 | 阿里巴巴集团控股有限公司 | 查询词推荐方法及装置 |
CN103971170A (zh) * | 2014-04-17 | 2014-08-06 | 北京百度网讯科技有限公司 | 一种用于对特征信息的变化进行预测的方法和装置 |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6611726B1 (en) | 1999-09-17 | 2003-08-26 | Carl E. Crosswhite | Method for determining optimal time series forecasting parameters |
JP4280045B2 (ja) * | 2002-09-04 | 2009-06-17 | 株式会社資生堂 | 生産量算定方法,生産量算定装置,生産量算定システム,生産量算定プログラムおよび記録媒体 |
KR100458459B1 (ko) * | 2004-01-27 | 2004-11-26 | 엔에이치엔(주) | 검색자의 검색 요청에 응답하여 검색 결과 목록을생성하고 검색어 광고를 제공하는 방법 및 검색어 광고제공 시스템 |
JP2007052533A (ja) * | 2005-08-16 | 2007-03-01 | Ntt Data Corp | 統計最適化統合装置、および統計最適化統合プログラム |
US7752190B2 (en) * | 2005-12-21 | 2010-07-06 | Ebay Inc. | Computer-implemented method and system for managing keyword bidding prices |
US8788306B2 (en) * | 2007-03-05 | 2014-07-22 | International Business Machines Corporation | Updating a forecast model |
CN101082972A (zh) * | 2007-05-30 | 2007-12-05 | 华为技术有限公司 | 预测用户对商品的兴趣的方法、装置和广告发布方法 |
JP5104567B2 (ja) * | 2007-12-21 | 2012-12-19 | 富士電機株式会社 | エネルギー需要予測装置 |
US20100094673A1 (en) * | 2008-10-14 | 2010-04-15 | Ebay Inc. | Computer-implemented method and system for keyword bidding |
US8265989B2 (en) * | 2009-05-05 | 2012-09-11 | The Nielsen Company, LLC | Methods and apparatus to determine effects of promotional activity on sales |
JP5345027B2 (ja) * | 2009-09-04 | 2013-11-20 | ウェザー・サービス株式会社 | 環境情報提供装置、システム、方法およびプログラム |
US8499066B1 (en) * | 2010-11-19 | 2013-07-30 | Amazon Technologies, Inc. | Predicting long-term computing resource usage |
CN102479190A (zh) * | 2010-11-22 | 2012-05-30 | 阿里巴巴集团控股有限公司 | 一种搜索关键词的估计值预测方法和装置 |
JP5678620B2 (ja) * | 2010-12-03 | 2015-03-04 | 株式会社日立製作所 | データ処理方法、データ処理システム、及びデータ処理装置 |
US10127568B2 (en) * | 2011-04-04 | 2018-11-13 | The Catholic University Of America | Systems and methods for improving the accuracy of day-ahead load forecasts on an electric utility grid |
-
2014
- 2014-04-17 CN CN201410154993.8A patent/CN103971170B/zh active Active
- 2014-12-16 EP EP14889627.7A patent/EP3133537A4/en not_active Ceased
- 2014-12-16 US US14/902,303 patent/US10474957B2/en active Active
- 2014-12-16 JP JP2016550932A patent/JP6254712B2/ja active Active
- 2014-12-16 WO PCT/CN2014/093952 patent/WO2015158149A1/zh active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101673385A (zh) * | 2009-09-28 | 2010-03-17 | 百度在线网络技术(北京)有限公司 | 消费预估方法和装置 |
CN103729351A (zh) * | 2012-10-10 | 2014-04-16 | 阿里巴巴集团控股有限公司 | 查询词推荐方法及装置 |
CN103617459A (zh) * | 2013-12-06 | 2014-03-05 | 李敬泉 | 一种多影响因素下商品需求信息预测方法 |
CN103971170A (zh) * | 2014-04-17 | 2014-08-06 | 北京百度网讯科技有限公司 | 一种用于对特征信息的变化进行预测的方法和装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP3133537A4 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107507030A (zh) * | 2017-08-22 | 2017-12-22 | 北京京东尚科信息技术有限公司 | 信息预测的方法和装置 |
Also Published As
Publication number | Publication date |
---|---|
US20160217383A1 (en) | 2016-07-28 |
JP6254712B2 (ja) | 2017-12-27 |
EP3133537A4 (en) | 2017-09-13 |
US10474957B2 (en) | 2019-11-12 |
JP2016540328A (ja) | 2016-12-22 |
CN103971170A (zh) | 2014-08-06 |
EP3133537A1 (en) | 2017-02-22 |
CN103971170B (zh) | 2017-09-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2015158149A1 (zh) | 一种用于对特征信息的变化进行预测的方法和装置 | |
US11170320B2 (en) | Updating machine learning models on edge servers | |
CN109697522B (zh) | 一种数据预测的方法和装置 | |
CN108605053B (zh) | 为未来动作优化用户界面数据缓存的方法、设备和系统 | |
TWI521453B (zh) | An estimation method and a device for estimating an estimated value of a keyword | |
US11018967B2 (en) | Determining an end user experience score based on client device, network, server device, and application metrics | |
JP6522160B2 (ja) | 情報配信方法、ならびに装置、サーバ、および記憶媒体 | |
US10657559B2 (en) | Generating and utilizing a conversational index for marketing campaigns | |
WO2015194182A1 (ja) | サービスチェーン管理装置、サービスチェーン管理システム、サービスチェーン管理方法、及び、プログラム記録媒体 | |
CN111247782B (zh) | 用于自动创建即时ad-hoc日历事件的方法和系统 | |
CN110147514B (zh) | 一种资源展示方法、装置及其设备 | |
WO2015127884A1 (en) | Method, device, system for displaying media data | |
CN110910201B (zh) | 信息推荐的控制方法、装置、计算机设备及存储介质 | |
JP6946082B2 (ja) | 広告配信支援装置、広告配信支援方法、およびプログラム | |
JP2016524227A (ja) | アプリケーション・ランキング算出装置および利用情報収集装置 | |
US11669762B2 (en) | Apparatus and method for forecasted performance level adjustment and modification | |
US20180341873A1 (en) | Adaptive prior selection in online experiments | |
CN105610886B (zh) | 信息推送的控制方法及信息推送平台 | |
JP5933487B2 (ja) | QoE推定装置、QoE推定方法及びプログラム | |
CN114430504B (zh) | 一种媒体内容的推荐方法以及相关装置 | |
JP6956564B2 (ja) | 情報処理装置、情報処理方法、およびプログラム | |
CN113326436B (zh) | 确定推荐资源的方法、装置、电子设备和存储介质 | |
CN109669779B (zh) | 用于确定数据的清理路径、清理数据的方法和设备 | |
JP7469452B1 (ja) | レコメンド決定装置、レコメンド決定方法、およびプログラム | |
US20210334734A1 (en) | Method and system for presenting digital task implemented in computer-implemented crowd-sourced environment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 14889627 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016550932 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14902303 Country of ref document: US |
|
REEP | Request for entry into the european phase |
Ref document number: 2014889627 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2014889627 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |