A data cinema profiling method for autonomous machine learning algorithms

Abstract

The present invention is a data cinema portrait technology for autonomous machine learning algorithms. It uses an API interface to pull together the basic data of all cinemas from major ticketing platforms, and uses data governance and control tools to organize data labels and trace associations for existing cinema basic data to obtain a cinema basic data asset catalog. A machine learning algorithm is introduced to select data features, and the data portrait model algorithm is trained based on the data features. The pros and cons of each cinema are presented in the form of a score. Based on the cinema portrait, it can fully reflect the scale, operating conditions, and development trends of the cinema, provide a basis for management personnel in cinema analysis, marketing promotion, risk prevention, and regulatory services, and realize the transformation of cinema management from extensive to precise, from passive to active, and from single to comprehensive and multi-level.

Description

Technical Field

[0001] The present invention relates to the field of machine learning, and in particular to a data cinema profiling method oriented to an autonomous machine learning algorithm. Background Art

[0002] Machine learning is a technology that simulates human learning behavior to acquire skills. It has been widely used in many industries such as finance, real estate, people's livelihood and archaeology, providing effective impetus for social development and innovation, and promoting the integration of information and industrialization development strategies.

[0003] With the improvement of living standards, the box office output value of China's film industry has continued to increase over the past few decades of development, but problems in cinema management have gradually emerged, including the large information time difference, poor marketing effect, weak investment promotion, and weak business management of domestic cinema chains. The reason is that there are problems in cinema management. The traditional extensive and passive traditional management methods lack scientific management methods and cannot comprehensively coordinate various management aspects, which urgently needs innovation and reform. Therefore, a data cinema portrait method for autonomous machine learning algorithms is needed. Summary of the invention

[0004] The purpose of the present invention is to solve the deficiencies of the prior art and provide a data cinema profiling method oriented to an autonomous machine learning algorithm.

[0005] In order to solve the above problems, the present invention adopts the following technical solutions:

[0006] A data cinema profiling method for an autonomous machine learning algorithm comprises the following steps:

[0007] Step 1: Obtain the factor data of the cinema through the cinema front desk system; the cinema front desk system is used to fill in the input data; obtain the basic data of the cinema through the transmission interface; obtain the business data of the cinema through the ETL tool;

[0008] Step 2: Input the data obtained in step 1 into the set image calculation application software to obtain the data cinema image, and end the step;

[0009] The acquisition of the image calculation application software in step 2 includes the following steps:

[0010] Step 21: Obtain the historical basic data of all theaters on major ticketing platforms through the transmission interface; extract the historical business data of all theaters in the existing system through the ETL tool;

[0011] Step 22: Obtain historical factor data of all cinemas through the cinema front desk system;

[0012] Step 23: Store the data obtained in steps 21 and 22 into an existing database and perform ETL processing to obtain operational data and complete storage;

[0013] Step 24: Use data governance and control tools to set data tag categories; label operational data according to data tag categories; set up associated traceability for associated data to form a data asset catalog and store it in the database; the data asset catalog includes basic indicators;

[0014] Step 25: Use machine learning algorithms to perform binary and multi-classification on the basic indicators in the data asset catalog, and obtain the data basic indicator features and store them in the database;

[0015] Step 26: Based on the characteristics of the basic indicators of the data and the historical business data of the cinema, the basic indicators are integrated and weighted to obtain a preliminary portrait calculation model;

[0016] Step 27: Set a time interval and repeat steps 21-26, continuously use the machine learning algorithm to optimize the data basic indicator characteristics, and adjust the weights of the data basic indicator characteristics accordingly to obtain the portrait calculation adjustment model;

[0017] Step 28: Develop the portrait calculation model into application software through development tools, and end the step.

[0018] Furthermore, the theater factor data in step 1 represents environmental factors around the theater, including local policies of the theater, supermarkets near the theater, transportation near the theater, and the proportion of theater audience types.

[0019] Furthermore, the transmission interface in step 1 adopts an API interface.

[0020] Furthermore, the ETL tools include RestCloud tools, informatica tools and kettle tools.

[0021] Furthermore, the basic data of the cinema in step 1 represents the fixed attributes of the cinema, including the cinema location, cinema business hours, cinema positioning, prime time for watching movies, number of theaters, theater types, and number of seats in theaters.

[0022] Furthermore, the cinema business data in step 1 represents business-related data, including box office data, merchandise data, investment promotion data, and activity data for each theater.

[0023] Furthermore, the labeling process of the operational data in step 24 means that a corresponding category identifier is assigned to each operational data according to a set condition.

[0024] Furthermore, the association tracing of the association data is obtained through the association degree between the data tag classes; wherein the association degree between the data tag classes is preset in the process of setting the data tag classes.

[0025] Furthermore, the machine learning algorithm in step 25 includes a Logistic Regression algorithm and a SVM algorithm.

[0026] Furthermore, the integration and weighting of basic indicators in step 26 includes the following steps:

[0027] Step 261: Integrate each indicator upward according to the label and the degree of relevance to form a third-level indicator, a second-level indicator, and a first-level indicator;

[0028] Step 262: assigning a score to each third-level indicator in combination with the historical business data of the cinema;

[0029] Step 263: Obtain the highest score among the third-level indicators corresponding to each second-level indicator, and add them up to obtain the full score, and then end the step.

[0030] The beneficial effects of the present invention are:

[0031] The portrait calculation application software obtained through machine learning algorithms completes the analysis of the number of cinema factors, cinema basic data and cinema business data, obtains the score of the cinema portrait, intuitively displays various factors of the cinema, and makes objective scores to help managers understand and manage the cinema;

[0032] By adopting a multi-level indicator method, the influencing factors of the cinema are classified and integrated according to labels and relevance, so that the final total score can objectively and accurately reflect the cinema's application situation;

[0033] By adopting the Logistic Regression algorithm and SVM algorithm, the classification of basic indicators is made more reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a schematic block diagram of the implementation of the digital cinema portrait in the first embodiment of the present invention;

[0035] Figure 2 This is a functional implementation diagram of the machine learning algorithm in the model according to the first embodiment of the present invention;

[0036] Figure 3 This is a diagram for implementing the index merging and weighting of the first embodiment of the present invention. DETAILED DESCRIPTION

[0037] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0038] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0039] Embodiment 1:

[0040] A data cinema profiling method for an autonomous machine learning algorithm comprises the following steps:

[0041] Step 1: Obtain the factor data of the cinema through the cinema front desk system; the cinema front desk system is used to fill in the input data; obtain the basic data of the cinema through the transmission interface; obtain the business data of the cinema through the ETL tool;

[0042] Step 2: Input the data obtained in step 1 into the set image calculation application software to obtain the data cinema image, and end the step.

[0043] The theater factor data in step 1 represents the environmental factors around the theater, including local policies of the theater, supermarkets near the theater, transportation near the theater, and the proportion of theater audience types in this example. The transmission interface uses an API interface; the ETL tools include RestCloud tools, informatica tools, and kettle tools, and the kettle tool is used in this example. The basic theater data represents the fixed attributes of the theater, including the theater location, theater business hours, theater positioning, prime time for watching movies, the number of theaters, theater types, and the number of theater seats in this example; the theater business data represents business-related data, including box office data, merchandise data, investment data, and activity data of branch halls and venues in this example. The theater front desk system is set up because the theater factor data cannot be automatically obtained through software or systems, so a data input system needs to be set up.

[0044] like Figure 1 As shown, the acquisition of the portrait calculation application software in step 2 includes the following steps:

[0045] Step 21: Obtain the historical basic data of all theaters on major ticketing platforms through the transmission interface; extract the historical business data of all theaters in the existing system through the ETL tool;

[0046] Step 22: Obtain historical factor data of all cinemas through the cinema front desk system;

[0047] Step 23: Store the data obtained in steps 21 and 22 into an existing database and perform ETL processing to obtain operational data and complete storage;

[0048] Step 24: Use data governance and control tools to set data tag categories; label operational data according to data tag categories; set up associated traceability for associated data to form a data asset catalog and store it in the database; the data asset catalog includes basic indicators;

[0049] Step 25: Introduce machine learning algorithms to perform binary and multi-classification on the basic indicators in the data asset catalog, obtain the data basic indicator features and store them in the database;

[0050] Step 26: Based on the characteristics of the basic indicators of the data and the historical business data of the cinema, the basic indicators are integrated and weighted to obtain a preliminary portrait calculation model;

[0051] Step 27: Set a time interval and repeat steps 21-26, continuously use the machine learning algorithm to optimize the data basic indicator characteristics, and adjust the weights of the data basic indicator characteristics accordingly to obtain the portrait calculation adjustment model;

[0052] Step 28: Develop the portrait calculation model into application software through development tools, and end the step.

[0053] The labeling of the operational data in step 24 means that the corresponding category identifier is assigned to each operational data through the set conditions. The association tracing of the associated data is obtained through the association degree between the data label classes; the association degree between the data label classes is preset in the process of setting the data label classes.

[0054] like Figure 2 As shown, the machine learning algorithm in step 25 includes the Logistic Regression algorithm and the SVM algorithm. The Logistic Regression algorithm is an existing algorithm for solving binary classification problems and identifying potential classifications and possibilities of data. The SVM algorithm is one of the existing machine learning methods for solving binary and multi-classification problems and ensuring rational classification.

[0055] like Figure 3As shown, the integration and weighting of basic indicators in step 26 include the following steps:

[0056] Step 261: Integrate each indicator upward according to the label and the degree of relevance to form a multi-level indicator; in this example, it is divided into three levels;

[0057] Step 262: assigning a score to each third-level indicator in combination with the historical business data of the cinema;

[0058] Step 263: Obtain the highest score among the third-level indicators corresponding to each second-level indicator, and add them up to obtain the full score, and then end the step.

[0059] The three levels of indicators in step 261 include primary indicators, secondary indicators and tertiary indicators, wherein the primary indicators are summary indicators, which in this example include five aspects: cinema strength, cinema competitiveness, cinema control, cinema credibility and cinema guidance. It should be noted that in some other implementations, other indicators can be set; the secondary indicators are subdivisions of the primary indicators, for example, in this example, the primary indicator is cinema strength, and the corresponding secondary indicators include cinema box office, etc.; the tertiary indicators are subdivisions of the links corresponding to the secondary indicators, for example, the secondary indicator is cinema box office, and the corresponding tertiary indicators are action film box office, science fiction film box office, suspense thriller film box office, etc.; points will be assigned to the tertiary indicators, for example, a box office of 10,000 or more in an action movie is assigned X points, a box office of 1,000 is assigned Y points, and a box office of 100 or less is assigned Z points.

[0060] The scoring of the three-level indicators in step 262 needs to refer to the historical cinema business data, wherein if the historical cinema business data is high, the score value of the indicator corresponding to the business data is high; on the contrary, if the historical cinema business data is low, the score value of the indicator corresponding to the business data is low. It should be noted that the scores of the three-level indicators can also be set in some other implementations.

[0061] The method for adjusting the weighted scoring in step 27 is to repeat steps 21-26, adjust the set weighted scoring, set the full score to 100 points, and proportionally reduce the scores of each third-level indicator under the three first-level indicators to obtain the final portrait calculation adjustment model. The full score is set to 100 points, which is convenient for understanding the high and low scores and for direct comparison.

[0062] The portrait calculation application software obtained in step 28 can display the total score of the cinema portrait, as well as the scores of each primary indicator, secondary indicator and tertiary indicator.

[0063] During the implementation process, the portrait calculation application software obtained through the machine learning algorithm completed the analysis of the number of cinema factors, cinema basic data and cinema business data, obtained the score of the cinema portrait, intuitively displayed the various factors of the cinema, and made objective scores to help managers understand and manage the cinema; by adopting a multi-level indicator method, the influencing factors of the cinema are classified and integrated according to labels and correlations, so that the final total score can objectively and accurately reflect the situation of the cinema; by adopting the Logistic Regression algorithm and the SVM algorithm, the classification of basic indicators is more reasonable.

[0064] The above description is only a specific example of the present invention and does not constitute any limitation to the present invention. It is obvious that for professionals in this field, after understanding the content and principle of the present invention, various modifications and changes in form and details may be made without departing from the principle and structure of the present invention, but these modifications and changes based on the idea of ​​the present invention are still within the scope of protection of the claims of the present invention.

Claims

1. A data cinema profiling method for autonomous machine learning algorithms, characterized in that: The steps include: Step 1: Obtain the factor data of the cinema through the cinema front desk system; the cinema front desk system is used to fill in the input data; obtain the basic data of the cinema through the transmission interface; obtain the business data of the cinema through the ETL tool; Step 2: Input the data obtained in step 1 into the set image calculation application software to obtain the data cinema image, and end the step; The acquisition of the image calculation application software in step 2 includes the following steps: Step 21: Obtain the historical basic data of all theaters on major ticketing platforms through the transmission interface; extract the historical business data of all theaters in the existing system through the ETL tool; Step 22: Obtain historical factor data of all cinemas through the cinema front desk system; Step 23: Store the data obtained in steps 21 and 22 into an existing database and perform ETL processing to obtain operational data and complete storage; Step 24: Use data governance and control tools to set data tag categories; label operational data according to data tag categories; set up associated traceability for associated data to form a data asset catalog and store it in the database; the data asset catalog includes basic indicators; Step 25: Use machine learning algorithms to perform binary and multi-classification on the basic indicators in the data asset catalog, and obtain the data basic indicator features and store them in the database; Step 26: Based on the characteristics of the basic indicators of the data and the historical business data of the cinema, the basic indicators are integrated and weighted to obtain a preliminary portrait calculation model; Step 27: Set a time interval and repeat steps 21-26, continuously use the machine learning algorithm to optimize the data basic indicator characteristics, and adjust the weights of the data basic indicator characteristics accordingly to obtain the portrait calculation adjustment model; Step 28: Develop the image calculation model into application software through development tools, and end the step; The theater factor data in step 1 represents the environmental factors around the theater, including local policies of the theater, supermarkets near the theater, transportation near the theater, and the proportion of theater audience types; The ETL tools include RestCloud tools, informatica tools and kettle tools; The basic data of the cinema in step 1 represents the fixed attributes of the cinema, including the location of the cinema, the operating hours of the cinema, the positioning of the cinema, the prime time for watching movies, the number of theaters, the type of theaters, and the number of seats in the theaters; The machine learning algorithm in step 25 includes a Logistic Regression algorithm and a SVM algorithm; The integration and weighting of basic indicators in step 26 include the following steps: Step 261: Integrate each indicator upward according to the label and the degree of relevance to form a third-level indicator, a second-level indicator, and a first-level indicator; Step 262: assigning a score to each third-level indicator in combination with the historical business data of the cinema; Step 263: Obtain the highest score among the third-level indicators corresponding to each second-level indicator, and add them up to obtain the full score, and then end the step.

2. The data cinema profiling method for autonomous machine learning algorithm according to claim 1, characterized in that: The transmission interface in step 1 adopts an API interface.

3. The data cinema profiling method for autonomous machine learning algorithm according to claim 1, characterized in that: The cinema business data in step 1 refers to business-related data, including box office data, merchandise data, investment data, and activity data for each theater.

4. The data cinema profiling method for autonomous machine learning algorithm according to claim 1, characterized in that: The labeling process of the operational data in step 24 means allocating a corresponding category identifier to each operational data according to a set condition.

5. The data cinema profiling method for autonomous machine learning algorithm according to claim 1, characterized in that: The association tracing of the association data is obtained through the association degree between the data tag classes; wherein the association degree between the data tag classes is preset in the process of setting the data tag classes.

Images