WO2023005635A1 - Method and apparatus for generating information - Google Patents

Abstract

The present application discloses a method and apparatus for generating information. A specific embodiment of the method comprises: obtaining target time sequence data of associated information of a target article; on the basis of historical time sequence data of associated information of a category to which the target article belongs, generating seasonal strength change information between months and a first seasonal strength identifier of the target article; generating a second seasonal strength identifier on the basis of text description information corresponding to the target article; and in response to determining that both the first seasonal strength identifier and the second seasonal strength identifier indicate strong seasonality, generating prediction information of the target article on the basis of the target time sequence data and the seasonal strength change information between the months.

Description

Method and apparatus for generating information

This patent application claims the priority of the Chinese patent application with the application number 202110843892.1 and the title of the invention "method and device for generating information" filed on July 26, 2021, which is incorporated by reference in its entirety middle.

technical field

The present application relates to the technical field of computers, in particular to the technical field of warehouse management, and in particular to a method and device for generating information.

Background technique

According to the distribution and characteristics of the time series, the time series can be divided into multiple scenarios, and there are corresponding processing procedures and methods for the prediction of each scenario, and the prediction accuracy is different. Among them, there are great difficulties in the identification and prediction of time series with 'seasonality'. The difficulties mainly lie in the judgment of the strength of 'seasonality', cycle identification, data processing flow, and the way of model coupling.

In related technologies, there are generally only data processing and forecasting stages for 'seasonal' time series forecasting, and there are roughly three operations as follows: 1. Based on statistical thinking, split the time series into components; 2. Based on machine learning ideology, constructing the characteristics of the same period and recent period based on historical data; 3. The baseline forecast does not consider 'seasonality', and adopts the method of year-on-year comparison for post-processing. The above operations have the disadvantages of being easily affected by sensitive values, weak interpretability, and highly dependent on experimental results and data mining, which can easily lead to large deviations in prediction results and affect the effectiveness of warehouse management.

Contents of the invention

Embodiments of the present application provide a method, device, device, storage medium, and computer program product for generating information.

In one or more embodiments, the present application provides a method for generating information, the method comprising: acquiring target time series data of associated information of a target item; based on historical time series data of associated information of the category to which the target item belongs, generating The seasonal strength change information between the months of the target item and the first seasonal strength mark; based on the text description information corresponding to the target item, a second seasonal strength mark is generated; in response to determining the first seasonal strength Both the mark and the second seasonal strength mark indicate strong seasonality, and the prediction information of the target item is generated based on the target time series data and seasonal change information between months.

In one or more embodiments, the present application provides a device for generating information, which includes: a data acquisition module configured to acquire target time-series data of associated information of a target item; a first generation module configured Based on the historical time series data of the associated information of the category to which the target item belongs, generate the seasonal strength change information and the first seasonal strength identification of the target item between each month; the second generating module is configured to The text description information corresponding to the item generates a second seasonal strength indicator; the generating information module is configured to respond to determining that both the first seasonal strength indicator and the second seasonal intensity indicator indicate strong seasonality, based on the target timing The data and the seasonal strength change information between the months are used to generate the forecast information of the target item.

In one or more embodiments, the present application provides an electronic device, the electronic device includes one or more processors; a storage device, on which one or more programs are stored, when the one or more programs are The one or more processors are executed, so that the one or more processors implement the method for generating information according to any embodiment of the first aspect.

In one or more embodiments, the present application provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the method for generating information according to any embodiment of the first aspect is implemented .

In one or more embodiments, the present application provides a computer program product, which includes a computer program, and when executed by a processor, the computer program implements the method for generating information according to any embodiment of the first aspect.

It should be understood that what is described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will be easily understood from the following description.

Description of drawings

FIG. 1 shows an exemplary system architecture diagram in which one or more embodiments can be applied;

Figure 2 shows a flowchart of one embodiment of a method of generating information of one or more embodiments;

Fig. 3 shows a schematic diagram of an application scenario of a method for generating information in one or more embodiments;

Figure 4 shows a flowchart of another embodiment of a method of generating information of one or more embodiments;

Fig. 5 shows a schematic diagram of an embodiment of an apparatus for generating information of one or more embodiments;

Fig. 6 shows a schematic structural diagram of a computer system of a server in one or more embodiments.

Detailed ways

Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and they should be regarded as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

FIG. 1 shows an exemplary system architecture 100 to which embodiments of the method of generating information of the present application may be applied.

As shown in FIG. 1 , a system architecture 100 may include terminal devices 101 , 102 , 103 , a network 104 and a server 105 . The network 104 is used as a medium for providing communication links between the terminal devices 101 , 102 , 103 and the server 105 . Network 104 may include various connection types, such as wires, wireless communication links, or fiber optic cables, among others.

The terminal devices 101, 102, 103 interact with the server 105 via the network 104 to receive or send messages and the like. Various communication client applications, such as shopping applications and communication applications, can be installed on the terminal devices 101, 102, and 103.

The terminal devices 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices with display screens, including but not limited to mobile phones and notebook computers. When the terminal devices 101, 102, 103 are software, they can be installed in the electronic devices listed above. It can be implemented as a plurality of software or software modules (for example to provide a service for generating information), or as a single software or software module. No specific limitation is made here.

The server 105 can be a server that provides various services, for example, to obtain the target time series data of the associated information of the target item; based on the historical time series data of the associated information of the category to which the target item belongs, generate the seasonal strength of the target item between each month Change information and the first seasonal strength mark; generate a second seasonal strength mark based on the text description information corresponding to the target item; respond to determining that both the first seasonal strength mark and the second seasonal strength mark indicate a strong Seasonality, based on the target time series data and seasonal strength change information between months, generate forecast information of the target item.

It should be noted that the server 105 may be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When the server is software, it can be implemented as a plurality of software or software modules (for example, to provide services for generating information), or can be implemented as a single software or software module. No specific limitation is made here.

It should be pointed out that the method for generating information provided by the embodiments of the present disclosure may be executed by the server 105, or may be executed by the terminal devices 101, 102, 103, or the server 105 and the terminal devices 101, 102, 103 may cooperate with each other implement. Correspondingly, each part (such as each unit, subunit, module, and submodule) included in the apparatus for generating information can be all set in the server 105, or can be all set in the terminal equipment 101, 102, 103, or can be set separately in the server 105 and the terminal devices 101, 102, 103.

It should be understood that the numbers of terminal devices, networks and servers in Fig. 1 are only illustrative. According to the implementation needs, there can be any number of terminal devices, networks and servers.

This application obtains the target time-series data of the related information of the target item; based on the historical time-series data of the related information of the category to which the target item belongs, generates the seasonal strength change information and the first seasonal strength indicator of the target item between months ; Based on the text description information corresponding to the target item, generate a second seasonal strength indicator; in response to determining that both the first seasonal strength indicator and the second seasonal intensity indicator indicate strong seasonality, based on the target time series data and The seasonal strength change information between each month generates the forecast information of the target item, which helps to achieve accurate forecast based on partial time series, and at the same time has a certain interpretation of the forecast. Among them, the accuracy is specifically reflected in The prediction of magnitude and the prediction of the time node of magnitude change. Further, forecast information can also be used for inventory management to improve the effectiveness of inventory management.

FIG. 2 shows a flow diagram 200 of an embodiment illustrating a method of generating information of one or more embodiments. In this embodiment, the method for generating information includes the following steps:

Step 201, acquiring target time-series data of associated information of a target item.

In this embodiment, the execution subject (such as the server 105 or the terminal devices 101, 102, and 103 shown in FIG. 1 ) can obtain the target time series data of the associated information of the target item in a wired or wireless manner.

Here, the target item may be any item for which information prediction is to be performed. The associated information may be various information related to the above-mentioned target item, for example, price, sales volume, number of likes, reserve and so on.

Among them, the target time-series data is usually numerical data of the time-series type, and may also include external information data explaining changes in the time-series numerical data, and text description information.

Specifically, taking the e-commerce industry as an example, the target item is a mosquito net, and the associated information is the sales volume. The target time series data can be the sales volume of mosquito nets in a certain region within a certain period of time. In addition, the target time series data can also include The temperature data, inventory data, etc. of the mosquito net, as well as the text description of the mosquito net, such as brand, color, applicable population, etc.

Here, the wireless connection method may include but not limited to 3G/4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods known or developed in the future.

Step 202, based on the historical time-series data of the associated information of the category to which the target item belongs, generate monthly seasonal strength change information and a first seasonal strength indicator of the target item.

In this embodiment, the execution subject can directly generate the seasonal strength change information and the first seasonal strength of the target item according to the time series data of the associated information of the category to which the target item belongs and the preset data threshold. Identification; it is also possible to generate at least one indicator data according to the time series data of the associated information of the category to which the target item belongs, and then based on the indicator data, generate the seasonal strength change information and the first seasonal strength identification of the target item between months .

Among them, the seasonal strength change information between the months of the target item is used to indicate the change information of the degree of association between the associated information of the target item in each month and the season, for example, the degree of association between the associated information from February to May and the season The degree of correlation between the associated information and the season is higher than that of the remaining months of the year. The first seasonal strength indicator is used to indicate whether the historical time series data of the associated information of the category to which the target item belongs is a strong seasonality, that is, whether it is strongly associated with the season.

Here, the first seasonal strength indicator may be represented by numbers, words, characters, etc., which is not limited in this application. For example, strong seasonality is "1" and weak seasonality is "0".

Step 203: Generate a second seasonal strength indicator based on the text description information corresponding to the target item.

In this embodiment, the execution subject can obtain the text description information corresponding to the target item, and extract the keyword information associated with the time series, for example, name: spring, summer, autumn, winter, attribute thickness, special holidays: lunar calendar, Gregorian calendar For festivals, etc., generate a second seasonal strength indicator according to the extracted keyword information.

If the keyword information associated with time series is not extracted from the text description information, the existing time series portrait pool can be used to identify the attribute similarity (such as metric learning) and browse similarity association (such as item embedding) of the text description information. ), etc., to generate the second seasonal strength indicator.

Here, the second seasonal strength flag is used to indicate whether the historical time series data of the associated information of the category to which the target item belongs is a strong seasonality, that is, whether it is strongly correlated with the season. The second seasonal strength indicator can be represented by numbers, words, characters, etc., which is not limited in this application. For example, strong seasonality is "1" and weak seasonality is "0".

In addition, it should be pointed out that if the keyword extraction and the above-mentioned similarity identification are not enough to generate the second seasonal strength indicator, it can be determined that the target time series data is not enough to support the seasonality identification and can be ignored.

Step 204, in response to determining that both the first seasonal strength flag and the second seasonal strength flag indicate strong seasonality, generate forecast information of the target item based on the target time series data and seasonal strength change information between months .

In this embodiment, after obtaining the first seasonal strength flag and the second seasonal strength flag, the executive body judges the first seasonal strength flag and the second seasonal strength flag, if both If both indicate strong seasonality, the time series data of the associated information of the target item in the future preset time period can be generated according to the target time series data and the seasonal strength change information between the above months.

Specifically, the target item is a mosquito net, the related information of the target item is the sales volume, and the time series data of the related information of the target item is the daily sales volume of the mosquito net during 2020-5-1~2020-5-5, if the first seasonality is strong or weak Both the mark and the second seasonal strength mark indicate that the target time-series data is strong seasonality, then the sales data of the target item mosquito net in the next year can be generated according to the target time-series data and the seasonal strength change information between months.

In addition, after the execution subject generates the prediction information, it can also keep the prediction information on the distributed data storage, and push it to the database through the plumber data, and display it in the downstream system and the front end in two ways: Hive table and Mysql.

In some optional manners, generating forecast information of the target item based on the target time-series data and seasonal strength change information between months includes: in response to determining that the time-series length of the target time-series data is less than a preset length threshold, based on The target time-series data, the historical time-series data of the related information of the category to which the target item belongs, and the seasonal strength change information between each month generate the forecast information of the target item.

In this implementation, the execution subject judges the time-series length of the target time-series data. If the time-series length of the target time-series data is less than the preset length threshold, the execution subject can first perform baseline prediction based on the target time-series data to obtain the baseline prediction result, and then The baseline prediction results are superimposed on the historical time series data of the related information of the category of the target item to obtain the superimposed prediction results, and further combined with the seasonal strength change information between months (such as magnitude rising and falling inflection points) to generate prediction information for the target item.

Wherein, the baseline prediction may include multiple types, for example, a baseline prediction based on statistical learning, a baseline prediction based on machine learning, a baseline prediction based on an ensemble mechanism, and the like.

Here, the preset length threshold can be set according to experience and actual needs, such as one year, half a year, etc., which is not limited in this application.

In this implementation, in response to determining that the time series length of the target time series data is less than the preset length threshold, the target time series data is generated based on the target time series data, the historical time series data of the associated information of the category to which the target item belongs, and the seasonal strength change information between months. The forecast information of the item is helpful to realize the short-term seasonal forecast, and at the same time, it implies the inflection point of the rise and fall of the magnitude.

In some optional manners, generating forecast information of the target item based on the target time-series data and seasonal change information between months includes: in response to determining that the time-series length of the target time-series data is greater than or equal to a preset length threshold, The target time-series data is factorized to obtain the sub-target time-series data corresponding to the seasonal factor; based on the sub-target time-series data and the seasonal strength change information between months, the forecast information of the target item is generated.

In this implementation, the execution subject can judge the time series length of the target time series data. If the time series length of the target time series data is greater than or equal to the preset length threshold, the execution subject can factorize the target time series data to obtain the corresponding seasonal factor The time-series data of the sub-targets, and based on the time-series data of the sub-targets, further combine the seasonal strength change information between months to generate the forecast information of the target items. Here, the executive body can use the decomposition method in the existing technology or future development technology, for example, X11 decomposition method, SEATS (Seasonal Extraction in ARIMA Time Series, seasonal extraction in ARIMA time series) decomposition method, etc., to analyze the target time series data Factor disassembly, that is, seasonal time series decomposition, that is, assuming that the target time series data is an additive model to split the time series into multiple factors, and obtain the sub-target time series data corresponding to the seasonal factors.

It should be pointed out that the executive body can also further combine seasonally related factors, such as weather, festivals, etc., on the basis of sub-target time series data and seasonal change information between months to generate forecasts of target items information.

In this implementation method, in response to determining that the time series length of the target time series data is greater than or equal to the preset length threshold, the target time series data is factorized to obtain the sub-target time series data corresponding to the seasonal factor; The seasonal strength change information of the target item is generated, which helps to realize the long-term seasonal forecast, and at the same time implies the inflection point of the rise and fall of the magnitude.

In some optional ways, based on the target time series data and the seasonal strength change information between months, the forecast information of the target item is generated, including: smoothing the target time series data to obtain the smoothed target time series data ; Generate forecast information of the target item based on the smoothed target time series data and the seasonal strength change information between the months.

In this implementation, before performing information prediction, the execution subject needs to perform smoothing processing on the target time-series data according to the auxiliary data to obtain the smoothed target time-series data. The amplitude of the target time series data is enhanced or weakened, and then the forecast information of the target item is generated according to the smoothed target time series data and seasonal strength change information between months.

Specifically, period 1 and period 2 in the target time-series data are the same period of two consecutive years, and the magnitude of the associated information value of period 1 is very low, for example, 0, but combined with auxiliary data, low magnitude does not In line with the actual situation, for example, due to reasons such as failure to sell, it is necessary to further combine the seasonal strength information of each month of the target item, page views and other auxiliary data to perform amplitude enhancement processing on the data in period 1, so that the processed target time series The data is more in line with objective facts, and the reason for the abnormality in period 1 will not be used as a reference factor in subsequent predictions.

This implementation method smoothes the target time-series data to obtain the smoothed target time-series data; based on the smoothed target time-series data and the seasonal strength change information between months, the forecast information of the target item is generated, and further The reliability and rationality of the generated forecast information are improved.

In some optional manners, the method further includes: in response to determining that there is a factor that affects the associated information of the time period corresponding to the prediction information, adjusting the prediction information based on the factor.

In this implementation, after obtaining the forecast information, if the execution subject determines that there are known factors or potential factors that affect the associated information of the time period corresponding to the forecast information, the forecast information is adjusted according to the factors.

Specifically, the forecast information is the monthly sales of mosquito nets in the next year. The executive body can use the factors that have recently occurred to affect the sales of subsequent mosquito nets, or the known factors that will affect the sales of mosquito nets that will occur in a certain time period in the next year, The forecast information is adjusted to obtain the adjusted forecast information.

In this implementation, the accuracy and reliability of the prediction information are further improved by adjusting the prediction information based on the factors in response to determining that there are factors that affect the associated information of the time period corresponding to the prediction information.

In some optional manners, the method further includes: adjusting the inventory information of the target item based on the forecast information of the target item.

In this implementation, after obtaining the forecast information of the target item, the execution subject may further increase or decrease the inventory of the target item according to the forecast information.

This implementation method adjusts the inventory information of the target item based on the forecast information of the target item, that is, adjusts the inventory according to the magnitude of the target item in the seasonal cycle indicated by the forecast information to optimize inventory management, that is, according to the preset time length in advance , to predict the magnitude trend of the target item given, and make pre-preparation to save a lot of cost and improve efficiency.

Continuing to refer to FIG. 3 , FIG. 3 is a schematic diagram illustrating an application scenario of the method for generating information in one or more embodiments.

In the application scenario in Figure 3, the execution subject 301 obtains the relevant information of the target item, such as mosquito nets, such as the sales volume, and the target time series data 302, for example, the mosquito nets are sold every day from 2020-5-1 to 2020-5-5 sales; based on the category of the target item, for example, home textiles, the historical time series data of the associated information, generate the seasonal strength change information 303 and the first seasonal strength indicator 304 between the months of the target item; based on the target item Corresponding text description information, generate a second seasonal strong and weak sign 305; in response to determining that both the first seasonal strong and weak sign 304 and the second seasonal strong and weak sign 305 indicate strong seasonality 306, based on the target time series data and each month The seasonal strength change information 307 generates forecast information of the target item, for example, the sales data of mosquito nets in the next year. Further, the execution subject 301 can optimize the inventory management information according to the above forecast information.

The method for generating information disclosed in this disclosure obtains the target time-series data of the related information of the target item; based on the historical time-series data of the related information of the category to which the target item belongs, generates the seasonal strength change information and the first month-to-month seasonal change information of the target item A seasonal strong and weak mark; based on the text description information corresponding to the target item, generate a second seasonal strong and weak mark; in response to determining that both the first seasonal strong and weak mark and the second seasonal strong and weak mark indicate strong seasonality, based on The target time-series data and the seasonal strength change information between months generate forecast information of the target item, which effectively improves the accuracy and interpretability of the forecast information.

Further referring to FIG. 4 , it shows a flow 400 of another embodiment of the method for generating information shown in FIG. 2 . In this embodiment, the process 400 of the method for generating information may include the following steps:

Step 401, acquiring target time-series data of associated information of the target item.

In this embodiment, for implementation details and technical effects of step 401, reference may be made to the description of step 201, and details are not repeated here.

Step 402: Divide the historical time-series data of the associated information of the category to which the target item belongs to the monthly dimension to obtain the monthly-dimension time-series data.

In this embodiment, after obtaining the historical time-series data of the associated information of the category to which the target item belongs, the execution subject can divide the time-series data into the monthly dimension to obtain the monthly-dimension time-series data.

Specifically, the target item is a mosquito net, the category of the target item is home textiles, the related information of the category of the target item is the sales volume of home textiles, and the historical time series data is the daily sales volume of home textiles in the past year. The executive body can summarize the historical time series data , to get the monthly dimension time series data, that is, the monthly sales data of home textiles in the past year.

Step 403: Generate at least one intensity index based on the ranking of the associated information values of the target item in the monthly dimension time series data in each month.

In this embodiment, the execution subject can generate at least one intensity index according to the ranking of the associated information values of each month in the monthly dimension time series data in the month dimension time series data, and the strength index is used to indicate the seasonal strength of the category to which the target item belongs Information, that is, the degree of correlation between the related information of the category of the target item and the season.

Wherein, the intensity index may include a basic index and an extended index.

Here, the basic index is usually generated based on the monthly dimension time series data whose time series length is less than or equal to one year, and the basic index can include at least one of the following: the first basic index, which is based on the value of the associated information corresponding to each month in the monthly dimension time series data. The small sorting is to determine the month corresponding to the first preset number of related information values in the top ranking; the second basic index is to sort the related information values corresponding to each month in the monthly dimension time series data from large to small, The determined month corresponding to the second preset number of related information values that are ranked lower; the third basic index, the sum of the first preset number of related information values that are ranked higher than the above and the value of related information in the monthly dimension time series data The ratio of the sum; the fourth basic index, the ratio of the sum of the second preset number of related information values ranked last above to the sum of the related information values in the monthly dimension time series data.

Wherein, the first preset number and the second preset number can be set according to experience and actual needs, for example, three, four, etc., which are not limited in this application. Here, the first preset number and the second preset number may be the same or different, which is not limited in this application.

The extended index can be obtained from the basic index. The stretch index may include at least one of the following: a first stretch index, a second stretch index, and a third stretch index. Wherein, the first extended index and the second extended index are usually generated based on the monthly dimension time series data whose time series length is longer than one year. The first extended indicator: If the monthly dimension time series data is multi-year data, divide the monthly dimension time series data into sub-month dimension time series data, each sub-month dimension time series data corresponds to the year, and calculate the numerical ranking of the associated information in each sub-month dimension time series data The intersection of the first preset number of months (the first basic index) (for example, the top-ranked months in 2018 are March, April, and May, and the top-ranked months in 2019 are April, May month, June, the intersection is April, May) or calculate the intersection of the second preset number of months (the second basic index) that ranks lower in the associated information values in the time-series data of each sub-month dimension; the second extended index: If the time-series data of the monthly dimension is multi-year data, the time-series data of the monthly dimension is divided into the time-series data of the sub-month dimension. The union of a preset number of months (the first basic index) (for example, the top-ranked months in 2018 are March, April, and May, and the top-ranked months in 2019 are April, May, and June , the union is March, April, May, and June) or calculate the union of the second preset number of months (the second basic index) that ranks lower in the associated information values in the sub-month dimension time series data. The third extension index: the difference between the third basic index and the fourth basic index, where the third basic index is the sum of the first preset number of related information values and the monthly The ratio of the sum of associated information values in the time-series data of the dimension, the fourth basic indicator is the sum of the second preset number of associated information values in the time-series data of the monthly dimension and the value of the associated information in the time-series data of the monthly dimension The ratio of the sum.

Among them, the first basic index/second basic index is used to describe the amplitude of each month or the absolute strength of seasonality; the third basic index/fourth basic index is used to describe the amplitude of each month or the relative strength of seasonality; The first extension index is used to describe the intensity of the overlapping strength of each month for many years; the second extension index is used to describe the strength of the confusion of the strength and weakness of each month for many years; the third extension index is used to describe the high and low amplitude or seasonal differences within the year Strength of.

Step 404, based on at least one intensity index and at least one preset intensity threshold, generate seasonal strength change information and a first seasonal strength indicator between months.

In this embodiment, the execution subject can generate each Seasonal strength change information between months and first seasonal strength indicator.

Specifically, the target item is a mosquito net, the category of the target item is home textiles, the related information of the category of the target item is sales, and the monthly dimension time series data is the sales of home textiles per month in the past year. The executive body obtains the first basic index, that is, April, May, June, and July for the sales top in April, and the second basic index, that is, September, October, November, and December for the sales bottom in April, and the third The basic indicator, that is, top sales in April accounted for 92.1% of the annual sales, and the fourth basic indicator, that is, the bottom April sales accounted for 6.4% of the annual sales. Based on the third basic indicator and the fourth basic indicator, the third extended indicator is generated. , that is, the difference between 92.1% and 6.4% is 85.7%. If the preset strength threshold is 75%, the first seasonal strong and weak flag can be output as a flag indicating strong seasonality, such as "1", and various Seasonal strong and weak change information between months, such as months with strong seasonality: April, May, June, July.

Step 405: Generate a second seasonal strength indicator based on the text description information corresponding to the target item.

In this embodiment, for implementation details and technical effects of step 405, reference may be made to the description of step 203, which will not be repeated here.

Step 406, in response to determining that both the first seasonal strength flag and the second seasonal strength flag indicate strong seasonality, generate forecast information of the target item based on the target time series data and seasonal strength change information between months .

In this embodiment, for implementation details and technical effects of step 406, reference may be made to the description of step 204, which will not be repeated here.

It can be seen from FIG. 4 that, compared with the embodiment corresponding to FIG. 2 , the process 400 of the method for generating information in this embodiment reflects that the historical time series data of the associated information of the category to which the target item belongs is divided into the monthly dimension, and the obtained Monthly dimension time series data, based on the ranking of the associated information value of the target item in the monthly dimension time series data, at least one strength index is generated, and based on at least one strength index and at least one preset strength threshold value, the data between each month is generated The seasonal change information of the seasonal intensity and the first seasonal intensity indicator, and then in response to determining that both the first seasonal intensity indicator and the second seasonal intensity indicator indicate strong seasonality, based on the target time series data and the The seasonal strength change information of the target item is generated to generate the forecast information of the target item, that is, through the design of a reasonable strength index, a more general and accurate identification system is output, which avoids excessive reliance on parameter tuning experience and the construction of complex feature pools And feature engineering, which improves the efficiency and rationality of the generated seasonal strength change information and the first seasonal strength label, thereby improving the efficiency and rationality of the generated forecast information, and improving the efficiency and rationality of inventory management .

Further referring to FIG. 5 , as an implementation of the methods shown in the above figures, an embodiment of a device for generating information in one or more embodiments is shown. This device embodiment is similar to the method embodiment shown in FIG. 2 Correspondingly, the device can be specifically applied to various electronic devices.

As shown in FIG. 5 , the apparatus 500 for generating information in this embodiment includes: a data acquisition module 501 , a first generation module 502 , a second generation module 503 and an information generation module 504 .

Wherein, the data acquisition module 501 may be configured to acquire the target time series data of the associated information of the target item.

The first generation module 502 may be configured to generate seasonal strength change information and a first seasonal strength indicator of the target item between months based on the historical time series data of the associated information of the category to which the target item belongs.

The second generation module 503 may be configured to generate a second seasonal strength indicator based on the text description information corresponding to the target item.

The generating information module 504 may be configured to respond to determining that both the first seasonal strong and weak indicator and the second seasonal strong and weak indicator indicate strong seasonality, based on the target time series data and the seasonality between the months change information of the strength of the target item, and generate prediction information of the target item.

In some optional forms of this embodiment, the first generation module further includes: a division data unit configured to divide the historical time series data of the associated information of the category to which the target item belongs to the monthly dimension to obtain the monthly dimension time series data; generate The index unit is configured to generate at least one strength index based on the ranking of the associated information value of the target item in each month in the monthly dimension time series data; the production identification unit is configured to generate at least one strength index based on at least one strength index and at least one preset Intensity threshold, generating seasonal strength change information between months and the first seasonal strength identification.

In some optional ways of this embodiment, the generating information module is further configured to: in response to determining that the time series length of the target time series data is less than a preset length threshold, The data and the seasonal strength change information between the months are used to generate the forecast information of the target item.

In some optional manners of this embodiment, the generating information module is further configured to: in response to determining that the time series length of the target time series data is greater than or equal to a preset length threshold, perform factor disassembly on the target time series data to obtain the corresponding seasonal factor The time-series data of the sub-targets; based on the time-series data of the sub-targets and the seasonal strength change information between the months, the prediction information of the target item is generated.

In some optional ways of this embodiment, the generating information module is further configured to: perform smoothing processing on the target time series data to obtain the smoothed target time series data; The seasonal strength change information of the target item is generated to generate forecast information.

In some optional manners of this embodiment, the apparatus further includes: an adjustment information module configured to adjust the prediction information based on the factors in response to determining that there is a factor that affects the associated information of the time period corresponding to the prediction information.

In some optional manners of this embodiment, the device further includes: an inventory adjustment module configured to adjust the inventory information of the target item based on the forecast information of the target item.

According to the embodiments of the present application, the present application also provides an electronic device and a readable storage medium.

As shown in FIG. 6 , it is a block diagram of an electronic device according to a method for generating information according to an embodiment of the present application.

600 is a block diagram of an electronic device according to the method for generating information according to the embodiment of the present application. Electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are by way of example only, and are not intended to limit implementations of the applications described and/or claimed herein.

As shown in FIG. 6, the electronic device includes: one or more processors 601, a memory 602, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are interconnected using different buses and can be mounted on a common motherboard or otherwise as desired. The processor may process instructions executed within the electronic device, including instructions stored in or on the memory, to display graphical information of a GUI on an external input/output device such as a display device coupled to an interface. In other implementations, multiple processors and/or multiple buses may be used with multiple memories and multiple memories, if desired. Likewise, multiple electronic devices may be connected, with each device providing some of the necessary operations (eg, as a server array, a set of blade servers, or a multi-processor system). In FIG. 6, a processor 601 is taken as an example.

The memory 602 is the non-transitory computer-readable storage medium provided in this application. Wherein, the memory stores instructions executable by at least one processor, so that the at least one processor executes the method for generating information provided in this application. The non-transitory computer-readable storage medium of the present application stores computer instructions, and the computer instructions are used to cause a computer to execute the method for generating information provided in the present application.

The memory 602, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs and modules, such as program instructions/modules corresponding to the method of generating information in the embodiment of the present application (for example, Acquiring data module 501, first generating module 502, second generating module 503 and generating information module 504 shown in Fig. 5). The processor 601 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 602, that is, implements the method of generating information in the above method embodiments.

The memory 602 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application program required by at least one function; the data storage area may store data created by use of an electronic device that generates information, and the like. In addition, the memory 602 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory 602 may optionally include memory that is remotely located relative to the processor 601, and these remote memories may be connected to electronic devices that generate information through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The electronic device of the method for generating information may further include: an input device 603 and an output device 604 . The processor 601, the memory 602, the input device 603, and the output device 604 may be connected through a bus or in other ways. In FIG. 6, connection through a bus is taken as an example.

The input device 603 can receive input digital or character information, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick and other input devices. The output device 604 may include a display device, an auxiliary lighting device (eg, LED), a tactile feedback device (eg, a vibration motor), and the like. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some implementations, the display device may be a touch screen.

Various implementations of the systems and techniques described herein can be implemented in digital electronic circuitry, integrated circuit systems, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpreted on a programmable system including at least one programmable processor, the programmable processor Can be special-purpose or general-purpose programmable processor, can receive data and instruction from storage system, at least one input device, and at least one output device, and transmit data and instruction to this storage system, this at least one input device, and this at least one output device an output device.

These computing programs (also referred to as programs, software, software applications, or codes) include machine instructions for a programmable processor and may be implemented using high-level procedural and/or object-oriented programming languages, and/or assembly/machine language calculation program. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or means for providing machine instructions and/or data to a programmable processor ( For example, magnetic disks, optical disks, memories, programmable logic devices (PLDs), including machine-readable media that receive machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.

To provide for interaction with the user, the systems and techniques described herein can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user. ); and a keyboard and pointing device (eg, a mouse or a trackball) through which a user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and can be in any form (including Acoustic input, speech input or, tactile input) to receive input from the user.

The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., as a a user computer having a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system can be interconnected by any form or medium of digital data communication, eg, a communication network. Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN) and the Internet.

A computer system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other.

According to the technical solutions of the embodiments of the present application, the accuracy and interpretability of generated information are effectively improved.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, the steps described in the present application may be executed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in the present application can be achieved, no limitation is imposed herein.

The above specific implementation methods are not intended to limit the protection scope of the present application. It should be apparent to those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made depending on design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application shall be included within the protection scope of this application.

Claims (17)

1. A method of generating information, the method comprising:

   Obtain the target time series data of the associated information of the target item;

   Based on the historical time-series data of the associated information of the category to which the target item belongs, generate seasonal strength change information and a first seasonal strength indicator of the target item between months;

   Generate a second seasonal strength indicator based on the text description information corresponding to the target item; and

   In response to determining that both the first seasonal intensity indicator and the second seasonal intensity indicator indicate strong seasonality, based on the target time series data and seasonal intensity change information between months, the Prediction information for the target item.
2. The method according to claim 1, wherein, based on the historical time series data of the associated information of the category to which the target item belongs, the seasonal strength change information and the first seasonal strength of the target item between each month are generated identification, including:

   Dividing the historical time series data of the associated information of the category to which the target item belongs to the monthly dimension to obtain the monthly dimension time series data;

   Based on the ranking of the associated information value of each month of the target item in the monthly dimension time series data, at least one strength index is generated, and the strength index is used to indicate the seasonal strength information of the category to which the target item belongs;

   Based on the at least one strength indicator and at least one preset strength threshold, seasonal strength change information between months and a first seasonal strength indicator are generated.
3. The method according to any one of claims 1-2, wherein the generating forecast information of the target item based on the target time series data and seasonal strength change information between months includes:

   In response to determining that the time-series length of the target time-series data is less than a preset length threshold, based on the target time-series data, historical time-series data of the associated information of the category to which the target item belongs, and seasonal strength changes between the months information to generate prediction information of the target item.
4. The method according to any one of claims 1-3, wherein the generating forecast information of the target item based on the target time-series data and seasonal strength change information between months includes:

   In response to determining that the time-series length of the target time-series data is greater than or equal to a preset length threshold, the target time-series data is factorized to obtain sub-target time-series data corresponding to seasonal factors; based on the sub-target time-series data and the monthly The seasonal strength change information among them is used to generate the forecast information of the target item.
5. The method according to any one of claims 1-4, wherein the generating forecast information of the target item based on the target time series data and seasonal strength change information between months includes:

   smoothing the target time-series data to obtain the smoothed target time-series data;

   The prediction information of the target item is generated based on the smoothed target time-series data and seasonal strength change information between months.
6. The method according to any one of claims 1-4, the method further comprising:

   In response to determining that there is a factor that affects the associated information of the time period corresponding to the forecast information, the forecast information is adjusted based on the factor.
7. The method according to any one of claims 1-4, the method further comprising:

   Based on the forecast information of the target item, the inventory information of the target item is adjusted.
8. An apparatus for generating information, the apparatus comprising:

   The data acquisition module is configured to acquire the target time series data of the associated information of the target item;

   The first generation module is configured to generate the seasonal strength change information and the first seasonal strength indicator of the target item between months based on the historical time series data of the associated information of the category to which the target item belongs;

   The second generation module is configured to generate a second seasonal strength indicator based on the text description information corresponding to the target item;

   A generating information module configured to respond to determining that both the first seasonal intensity indicator and the second seasonal intensity indicator indicate strong seasonality, based on the target time series data and seasonal intensity between months Weak change information to generate prediction information of the target item.
9. The device according to claim 8, wherein the first generating module further comprises:

   The dividing data unit is configured to divide the historical time-series data of the associated information of the category to which the target item belongs to the monthly dimension to obtain the monthly dimension time-series data;

   The generation indicator unit is configured to generate at least one intensity indicator based on the ranking of the associated information value of each month of the target item in the monthly dimension time series data, and the intensity indicator is used to indicate the seasonal intensity information of the category to which the target item belongs ;

   The production identification unit is configured to generate seasonal strength change information between months and a first seasonal strength indicator based on the at least one strength index and at least one preset strength threshold.
10. The device according to any one of claims 8-9, wherein the generating information module is further configured to:

    In response to determining that the time-series length of the target time-series data is less than a preset length threshold, based on the target time-series data, historical time-series data of the associated information of the category to which the target item belongs, and seasonal strength changes between the months information to generate prediction information of the target item.
11. The device according to any one of claims 8-10, wherein the generating information module is further configured to:

    In response to determining that the time-series length of the target time-series data is greater than or equal to a preset length threshold, the target time-series data is factorized to obtain sub-target time-series data corresponding to seasonal factors; based on the sub-target time-series data and the monthly The seasonal strength change information among them is used to generate the forecast information of the target item.
12. The device according to any one of claims 8-11, wherein the generating information module is further configured to:

    smoothing the target time-series data to obtain the smoothed target time-series data;

    The prediction information of the target item is generated based on the smoothed target time-series data and seasonal strength change information between months.
13. The device according to any one of claims 8-11, further comprising:

    The adjustment information module is configured to adjust the prediction information based on the factors in response to determining that there is a factor that affects the associated information of the time period corresponding to the prediction information.
14. The device according to any one of claims 8-11, further comprising:

    The inventory adjustment module is configured to adjust the inventory information of the target item based on the forecast information of the target item.
15. An electronic device comprising:

    at least one processor; and

    a memory communicatively coupled to the at least one processor; wherein,

    The memory stores information executable by the at least one processor, so that the at least one processor can execute the method according to any one of claims 1-7.
16. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the method according to any one of claims 1-7.
17. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-7.

Images