A smart box based on edge computing and data processing method thereof

Abstract

The present invention relates to the field of data processing technology, and is a smart box based on edge computing and a data processing method thereof, comprising: obtaining a historical search data set, using an edge computing unit to perform preprocessing operations on the historical search data set, obtaining an initial analysis data set, summarizing the initial analysis data set, obtaining a large data set, parsing the large data set, obtaining an optimized word frequency sequence, obtaining a sliding frequency extraction group, parsing the optimized word frequency sequence based on a search trend evaluation method and a sliding frequency extraction group, obtaining a time period film and television resource library, confirming that the time period film and television resource library has film and television resources corresponding to the search terms in the user's search catalog, initiating a new film and television reminder to the user, and completing the smart box data processing based on edge computing. The present invention can use edge computing to parse the search data of the smart box in the past time period to obtain a valuable time period film and television resource library, and provide intelligent new reminders to the user.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a smart box based on edge computing and a data processing method thereof. Background Art

[0002] The edge computing unit is a computing node deployed near the user, which has the ability to process and analyze data. It can receive information sent by the user, perform some preliminary processing, and send the processed data to the cloud to reduce the computing pressure of the data center. The smart box is a device used by users to connect the TV to the Internet.

[0003] In the prior art, when users use smart boxes to play TV, they often have the problem of not being able to find the film and television resources they want to watch. When introducing film and television resources to the smart box and its cloud, they only rely on the film and television market and its internal economic contracts to introduce film and television resources, for example, directly introducing popular film and television resources on the market.

[0004] Although the above technology can realize the introduction of film and television resources into the smart box, the smart box does not make reasonable use of the usage data left by the needs of this part of users. In addition, the users corresponding to the smart box are uneven, and the single measure of directly introducing popular film and television resources on the market cannot meet the needs of users using the smart box. Therefore, how to use user search data, meet user usage needs, and intelligently provide film and television resources is an urgent problem that needs to be solved. Summary of the invention

[0005] The present invention provides a smart box based on edge computing and a data processing method thereof. The computer can utilize edge computing to parse the search data of the smart box in the past time period to obtain a valuable time period film and television resource library, and provide users with intelligent new reminders.

[0006] To achieve the above objectives, the present invention provides a smart box based on edge computing and a data processing method thereof, comprising:

[0007] Obtain an edge computing-smart box analysis system, wherein the edge computing-smart box analysis system comprises: an edge computing unit, a data analysis unit and a smart box unit, wherein the smart box unit comprises a plurality of smart boxes;

[0008] Extract smart boxes from the smart box unit one by one, and perform the following operations on the extracted smart boxes:

[0009] Based on the smart box receiving the log authorization instruction from the user, obtaining the historical search data set of the smart box in the pre-constructed historical period according to the log authorization instruction, sending the historical search data set to the edge computing unit, and performing a preprocessing operation on the historical search data set by using the edge computing unit that receives the historical search data set to obtain an initial analysis data set;

[0010] Aggregate the initial analysis data sets corresponding to the multiple smart boxes to obtain a big data set, and use the data analysis unit to parse the big data set to obtain an optimized word frequency sequence, wherein the big data set includes multiple big data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in descending order of frequency;

[0011] Obtain a sliding frequency extraction group, parse the optimized word frequency sequence based on a pre-built search trend evaluation method and a sliding frequency extraction group, and obtain a plurality of valuable film and television terms, wherein the sliding frequency extraction group includes a preset number of blank groups;

[0012] Obtain a time period film and television resource library corresponding to a plurality of valuable film and television entries, and extract smart boxes from a plurality of smart boxes in sequence, intercept an initial analysis data set of the smart boxes based on a preset time period, obtain a time period search data set, parse the time period search data set, and obtain a user search directory, wherein the time period film and television resource library includes a plurality of film and television resources, and the film and television resources correspond to the valuable film and television entries one by one, and the user search directory includes zero, one or more search entries;

[0013] If it is confirmed that the film and television resources corresponding to the search terms in the user's search catalog exist in the film and television resource library of the time period, the smart box will be used to initiate a new film and television reminder based on the search terms to the user, and the smart box data processing based on edge computing will be completed based on the new film and television reminder.

[0014] Optionally, obtaining a historical search data set of the smart box in a pre-constructed historical period according to the log authorization instruction includes:

[0015] Parsing the log authorization instruction to obtain a historical period and a historical search log, wherein the historical search log includes a plurality of unit search records;

[0016] Extracting unit search records from the historical search log in sequence, and obtaining the time of the extracted unit search records to obtain the target time. If the target time is within the historical period, retain the extracted unit search records; otherwise, discard the extracted unit search records.

[0017] The retained unit search records are aggregated to obtain a historical search data set.

[0018] Optionally, the using the edge computing unit that receives the historical search data set to perform a preprocessing operation on the historical search data set to obtain an initial analysis data set includes:

[0019] The edge computing unit is used to extract historical search data from the historical search data set in sequence, and the following operations are performed on the extracted historical search data:

[0020] Extract time, information level and information content from historical search data to obtain multiple target information, and perform the following operations on target information in the multiple target information:

[0021] Determine whether the target information meets the pre-built field constraints. If the target information meets the field constraints, build a key identifier based on the target information, use the key identifier as the key and the target information as the value to build an initial information key-value pair. Otherwise, remove the historical search data corresponding to the target information.

[0022] After confirming that each of the multiple target information meets the corresponding field constraint condition, the initial information key-value pairs are summarized to obtain initial analysis data corresponding to the historical search data;

[0023] The initial analysis data are summarized to obtain the initial analysis data set.

[0024] Optionally, the utilizing a data analysis unit to parse a large data set to obtain an optimized word frequency sequence includes:

[0025] The following operations are performed on the big data in the big data collection:

[0026] Extracting information content from the big data to obtain content data, extracting search terms and search results from the content data, and if the search result is a preset no-result value, retaining the search term, otherwise, removing the big data corresponding to the content data;

[0027] The retained search terms are summarized to obtain an optimized term data set, and a recursive operation is performed on the optimized term data set to obtain an optimized term frequency sequence.

[0028] Optionally, performing a recursive operation on the optimized word data set to obtain an optimized word frequency sequence includes:

[0029] Extract optimized word data from the optimized word data set in sequence, and perform the following operations on the extracted optimized word data:

[0030] Obtaining an initial word frequency sequence, wherein the initial word frequency sequence includes a plurality of word frequency groups, and the word frequency group includes a target word and a frequency, and the initial value of the frequency is 0;

[0031] If there is a target word corresponding to the optimized word data in multiple word frequency groups in the initial word frequency sequence, then perform an increment operation on the frequency of the word frequency group corresponding to the target word in the initial word frequency sequence to obtain an updated word frequency group, use the updated word frequency group to update the initial word frequency sequence, and use the updated initial word frequency sequence as the initial word frequency sequence, and return to the step of sequentially extracting the optimized word data from the optimized word data set;

[0032] Otherwise, construct an initial added word frequency group with the optimized word data as the target word, and assign the frequency of the initial added word frequency group to 1 to obtain the added word frequency group, import the added word frequency group into the initial word frequency sequence, take the added word frequency group as the word frequency group, take the initial word frequency sequence after importing the added word frequency group as the initial word frequency sequence, and return to the step of extracting the optimized word data from the optimized word data set in sequence;

[0033] After confirming that the optimized word data in the optimized word data set have been extracted, a sorting operation is performed on the word frequency groups in the initial word frequency sequence in order of frequency from large to small to obtain an optimized word frequency sequence.

[0034] Optionally, the obtaining of the sliding frequency extraction group parses the optimized word frequency sequence based on a pre-built search trend evaluation method and the sliding frequency extraction group to obtain a plurality of valuable film and television terms, including:

[0035] Extract optimized word frequency groups from the optimized word frequency sequence in sequence, and map the extracted optimized word frequency groups to blank groups in the sliding frequency extraction group in sequence, until all blank groups in the sliding frequency extraction group are mapped with optimized word frequency groups, and obtain a term group to be analyzed, wherein the term group to be analyzed includes multiple term frequency groups to be analyzed, the term frequency groups to be analyzed correspond to the optimized word frequency groups one by one, and the multiple term frequency groups to be analyzed are sorted in the order from the first to the last in which the optimized word frequency groups are mapped to the blank groups;

[0036] Determine whether the term group to be analyzed is a pre-constructed value term group. If the term group to be analyzed is a value term group, then based on the term group to be analyzed and the sliding frequency extraction group, slide and intercept the optimized term frequency sequence to obtain an updated term group, and use the updated term group as the term group to be analyzed, and return to the step of determining whether the term group to be analyzed is a pre-constructed value term group until the term group to be analyzed is not a value term group, summarize the value term groups, obtain multiple value term groups, perform duplicate term deduplication operation on the multiple value term groups, and obtain multiple valuable film and television terms.

[0037] Optionally, the determining whether the to-be-analyzed term group is a pre-constructed value term group, if the to-be-analyzed term group is a value term group, includes:

[0038] Extract the word frequency groups to be analyzed from the word group to be analyzed in sequence, and perform the following operations on the extracted word frequency groups to be analyzed:

[0039] Obtain search terms corresponding to the word frequency group to be analyzed to obtain a target term, obtain a release date and a data cutoff date based on the target term, obtain a daily search frequency graph based on the release date, the data cutoff date and the target term, construct a daily frequency search volume regression model based on the daily search frequency graph and a pre-constructed frequency difference autoregression model, calculate the search volumes of multiple preset prediction target time points using the daily frequency search volume regression model, obtain multiple predicted search volumes, map the multiple predicted search volumes and the predicted target time points corresponding to each predicted search volume to a pre-constructed blank daily search frequency graph, and obtain a predicted frequency graph, wherein the horizontal axis of the predicted frequency graph is the predicted target time point, and the vertical axis is the predicted search volume, and the predicted target time points correspond to the predicted search volumes one by one;

[0040] Performing a linear fitting operation on the predicted frequency graph to obtain a fitting straight line, obtaining the slope of the fitting straight line, and obtaining a search frequency trend value. If the search frequency trend value is greater than a preset trend threshold, the target term is confirmed as a unit value term;

[0041] Determine whether the number of unit value terms in the term group to be analyzed is greater than a preset value number;

[0042] If the number of unit value terms in the term group to be analyzed is greater than the value number, the term group to be analyzed is confirmed to be a value term group.

[0043] Optionally, obtaining a daily search frequency graph based on the release date, the data cutoff date and the target term includes:

[0044] Obtaining a search period based on the release date and the data cutoff date, dividing the search period to obtain a day sequence, and constructing an initial day search frequency coordinate system based on the target term with the day sequence as the horizontal axis and the frequency as the vertical axis;

[0045] The day sequences are extracted from the day sequence in sequence, and the total daily search frequency corresponding to the target term is extracted from the big data set based on the extracted day sequences and the target term, and the total daily search frequency and the corresponding day sequence are mapped to the initial daily search frequency coordinate system. After confirming that all the day sequences in the day sequence have been extracted, the initial daily search frequency graph is obtained.

[0046] Optionally, the fitting straight line is as follows:

[0047] in, represents the predicted search volume in the fitted line, represents the predicted target time point in the fitting line, Indicates The index of the predicted target time point, Indicates the total number of predicted target time points, Indicates Predict the target time point, Represents the mean of multiple predicted target time points, Indicates Predict the predicted search volume corresponding to the target time point, Represents the mean of the predicted search volumes corresponding to multiple predicted target time points.

[0048] To achieve the above objectives, the present invention also provides a smart box based on edge computing and a data processing system thereof, comprising:

[0049] A system acquisition module, used to acquire an edge computing-smart box analysis system, wherein the edge computing-smart box analysis system includes: an edge computing unit, a data analysis unit and a smart box unit, wherein the smart box unit includes a plurality of smart boxes;

[0050] The initial analysis data set acquisition module is used to sequentially extract smart boxes from the smart box units, and perform the following operations on the extracted smart boxes: based on the smart box receiving a log authorization instruction from the user, obtain a historical search data set of the smart box in a pre-constructed historical period according to the log authorization instruction, send the historical search data set to the edge computing unit, and perform a preprocessing operation on the historical search data set using the edge computing unit that receives the historical search data set to obtain an initial analysis data set;

[0051] A valuable film and television entry acquisition module is used to aggregate the initial analysis data sets corresponding to multiple smart boxes to obtain a large data set, and use the data analysis unit to parse the large data set to obtain an optimized word frequency sequence, wherein the large data set includes multiple large data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in order from large to small in frequency, and a sliding frequency extraction group is obtained. The optimized word frequency sequence is parsed based on the pre-constructed search trend evaluation method and the sliding frequency extraction group to obtain multiple valuable film and television entries, wherein the sliding frequency extraction group includes a preset number of blank groups;

[0052] The smart box prompt module is used to obtain a time period film and television resource library corresponding to multiple valuable film and television terms, and extract smart boxes from multiple smart boxes in sequence, intercept the initial analysis data set of the smart box based on the preset time period, obtain the time period search data set, parse the time period search data set, and obtain the user search directory, wherein the time period film and television resource library includes multiple film and television resources, and the film and television resources correspond to the valuable film and television terms one by one, and the user search directory includes zero, one or more search terms. If it is confirmed that the film and television resources corresponding to the search terms in the user search directory exist in the time period film and television resource library, the smart box is used to initiate a new film and television prompt based on the search terms to the user, and the smart box data processing based on edge computing is completed based on the new film and television prompt.

[0053] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0054] A memory storing at least one instruction;

[0055] The processor executes the instructions stored in the memory to implement the above-mentioned edge computing-based smart box and data processing method thereof.

[0056] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned edge computing-based smart box and its data processing method.

[0057] In order to solve the problems described in the background technology, the present invention first obtains an edge computing-smart box analysis system, extracts smart boxes from the smart box units in sequence, and performs the following operations on the extracted smart boxes: based on the smart box receiving the log authorization instruction from the user, the historical search data set of the smart box in the pre-constructed historical period is obtained according to the log authorization instruction, the historical search data set is sent to the edge computing unit, and the edge computing unit that receives the historical search data set performs preprocessing operations on the historical search data set to obtain an initial analysis data set. In the present invention, when the smart box receives the log authorization instruction from the user, the smart box can upload and send the locally retained logs to the edge gateway and other places for processing, so as to provide intelligent services to the user with the user's permission. The log authorization instruction also ensures that the user's usage traces are not leaked. The present invention also uses the historical search data set to filter within the historical period to obtain the historical search data set within the historical period. Aggregate the initial analysis data sets corresponding to multiple smart boxes to obtain a big data set, use the data analysis unit to parse the big data set, and obtain an optimized word frequency sequence, wherein the big data set includes multiple big data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in order from large to small in frequency, obtain a sliding frequency extraction group, and parse the optimized word frequency sequence based on the pre-constructed search trend evaluation method and the sliding frequency extraction group to obtain multiple valuable film and television entries, wherein the sliding frequency extraction group includes a preset number of blank groups. The present invention provides data support and target guidance for filtering out film and television resources that users search for but cannot find in the smart box, thereby updating the TV box resource library in the cloud. The optimized word frequency sequence is evaluated using a sliding frequency extraction group, and a unit-value term is confirmed when the search frequency trend value is greater than a preset trend threshold. When the number of unit-value terms in the sliding frequency extraction group is greater than the preset value number, a value term group is obtained, and the optimized word frequency sequence is slidingly intercepted using the sliding frequency extraction group. When the term group to be analyzed is not a value term group, it means that part or all of the search volume of the lagging optimized word frequency group is not a unit-value term, and its search trend therefore stops the sliding interception operation at this time, saving computational effort while extracting valuable film and television terms as much as possible.Obtain a time period film and television resource library corresponding to multiple valuable film and television terms, and extract smart boxes from multiple smart boxes in sequence, intercept the initial analysis data set of the smart box based on the preset time period, obtain the time period search data set, parse the time period search data set, and obtain the user search directory, wherein the time period film and television resource library includes multiple film and television resources, and the film and television resources correspond to the valuable film and television terms one by one, and the user search directory includes zero, one or more search terms. If it is confirmed that the film and television resources corresponding to the search terms in the user search directory exist in the time period film and television resource library, then use the smart box to initiate a new film and television reminder based on the search terms to the user, and complete the smart box data processing based on edge computing based on the new film and television reminder. After the introduction of the time period film and television resource library, the present invention provides a new reminder for searching for the terms existing in the time period film and television resource library in a short period of time, thereby improving the intelligence of the smart box and the user experience of the user. Therefore, the present invention can use edge computing to parse the search data of the smart box in the past time period to obtain a valuable time period film and television resource library, and provide intelligent new reminders to the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A schematic diagram of a flow chart of a smart box based on edge computing and a data processing method thereof provided in one embodiment of the present invention;

[0059] Figure 2 A functional module diagram of a smart box based on edge computing and its data processing system provided by an embodiment of the present invention;

[0060] Figure 3 A schematic diagram of the structure of an electronic device for implementing the edge computing-based smart box and its data processing method provided by an embodiment of the present invention;

[0061] Description of reference numerals:

[0062] 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0063] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0064] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0065] The embodiment of the present application provides a smart box based on edge computing and a data processing method thereof. The execution subject of the smart box based on edge computing and the data processing method thereof includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the smart box based on edge computing and the data processing method thereof can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0066] Reference Figure 1 FIG. 1 is a flow chart of a smart box based on edge computing and a data processing method thereof provided by an embodiment of the present invention. In this embodiment, the smart box based on edge computing and a data processing method thereof include:

[0067] S1. Obtain an edge computing-smart box analysis system, wherein the edge computing-smart box analysis system includes: an edge computing unit, a data analysis unit and a smart box unit, wherein the smart box unit includes multiple smart boxes.

[0068] It is understandable that the edge computing-smart box analysis system is a system that integrates an edge computing unit, a data analysis unit, and a smart box unit, wherein the edge computing unit is a computing node deployed near the smart box unit, which has the ability to process and analyze data, can receive information sent by the smart box unit, and perform some preliminary processing, and send the preliminary processed information to the pre-built cloud to reduce the computing pressure of the data center. This technology is a prior art and will not be repeated here. The data analysis unit is a unit that analyzes and processes the data received by the edge computing unit. The smart box unit is a unit composed of multiple smart boxes, and the smart box is a device used by users to connect the TV to the Internet.

[0069] S2. Extract smart boxes from the smart box unit in sequence, and perform the following operations on the extracted smart boxes: based on the smart box receiving the log authorization instruction from the user, obtain the historical search data set of the smart box in the pre-constructed historical period according to the log authorization instruction.

[0070] It is understandable that the log authorization instruction is the instruction to which the user responds to authorize after the smart box initiates an authorization request to the user. When the smart box receives the log authorization instruction from the user, the smart box can upload and send the locally retained logs to the edge gateway for processing, so as to provide intelligent services to the user with the user's permission. The log authorization instruction also ensures that the user's usage traces are not leaked. The historical period is a period of time in the past, and its range can be limited artificially, for example, the past 30 days.

[0071] Further, the obtaining of a historical search data set of the smart box in a pre-constructed historical period according to the log authorization instruction includes:

[0072] Parsing the log authorization instruction to obtain a historical period and a historical search log, wherein the historical search log includes a plurality of unit search records;

[0073] Extracting unit search records from the historical search log in sequence, and obtaining the time of the extracted unit search records to obtain the target time. If the target time is within the historical period, retain the extracted unit search records; otherwise, discard the extracted unit search records.

[0074] The retained unit search records are aggregated to obtain a historical search data set.

[0075] It is understandable that the historical search log is the log of the smart box in the historical period. The log refers to the record generated when the user performs a search operation in the smart box. This technology is existing technology and will not be repeated here. The unit search record is the record of a user performing a search operation in the smart box. For example, the unit search record of a user in a search is [Movie A], the time is [a year b month c day d hour e minute f second], and the search result is [none].

[0076] Furthermore, the target time is the time in the unit search record. If the target time is within the historical period, it means that the target time is within the time interval formed by the historical period. For example, the historical period is [a year b month c day d hour e minute 0 second - a year b month g day d hour e minute 0 second], and the time of [c day] is earlier than [g day], and the target time is [a year b month c day d hour e minute f second], and f is greater than zero, then it is considered that the target time is within the historical period, and the unit search record is retained.

[0077] It is understandable that the purpose of this step is to filter the historical search data set within the historical period. In the embodiment of the present invention, the historical search data set is limited to not being an empty set, that is, if the smart box does not have a historical search log within the historical period, it is considered that the smart box has no need for data processing or analysis.

[0078] S3. Send the historical search data set to the edge computing unit, and use the edge computing unit that receives the historical search data set to perform a preprocessing operation on the historical search data set to obtain an initial analysis data set.

[0079] Further, the edge computing unit that receives the historical search data set performs a preprocessing operation on the historical search data set to obtain an initial analysis data set, including:

[0080] The edge computing unit is used to extract historical search data from the historical search data set in sequence, and the following operations are performed on the extracted historical search data:

[0081] Extract time, information level and information content from historical search data to obtain multiple target information, and perform the following operations on target information in the multiple target information:

[0082] Determine whether the target information meets the pre-built field constraints. If the target information meets the field constraints, build a key identifier based on the target information, use the key identifier as the key and the target information as the value to build an initial information key-value pair. Otherwise, remove the historical search data corresponding to the target information.

[0083] After confirming that each of the multiple target information meets the corresponding field constraint condition, the initial information key-value pairs are summarized to obtain initial analysis data corresponding to the historical search data;

[0084] The initial analysis data are summarized to obtain the initial analysis data set.

[0085] It is understandable that the information level is the severity or importance level of the historical search data, which is used to identify and filter the information in the log, including: DEBUG, INFO, WARN, etc. The information content is the search content and search results in the historical search data. For example, the unit search record of a user in a search is [Movie A], the time is [a year b month c day d hour e minute f second], the search result is [none], and the information level is [INFO]. The historical search data is: [abc d:e:f INFO Usersearched for 'Movie A', The search result is [none]], and the information content is: User searchedfor 'Movie A', The search result is [none]. The search content is [Movie A].

[0086] Furthermore, the field constraint is a constraint that the target information needs to satisfy to ensure the integrity of the target information, such as a non-empty constraint, a unique constraint, etc. This technology is a prior art. In the embodiment of the present invention, different target information needs to satisfy the corresponding field constraint. For example, time needs to satisfy the field constraint corresponding to time, such as a non-empty constraint, a unique constraint, and a field type constraint, and the field type constraint is a timestamp. If the target information is the information level, the corresponding field constraint needs to be satisfied, such as a non-empty constraint, a unique constraint, and a field type constraint, and the field type constraint is an enumeration. The enumeration only includes specific values, such as DEBUG, INFO, WARN, etc. If the target information is the information content, the corresponding field constraint needs to be satisfied, such as a non-empty constraint, a unique constraint, and a field type constraint, and the field type constraint is text.

[0087] It is understandable that the key identifier is an identifier used to construct a key-value pair for the target information. For example, if the target information is time, the key identifier is [time], and the target information is [abc d:e:f], then the initial information key-value pair is [time, abc d:e:f]. The initial analysis data is a set of initial information key-value pairs corresponding to the historical search data. Constructing the initial information key-value pair is a technology for constructing key-value pairs, which is a prior art and will not be repeated here.

[0088] S4. Aggregate the initial analysis data sets corresponding to multiple smart boxes to obtain a big data set, and use the data analysis unit to parse the big data set to obtain an optimized word frequency sequence, wherein the big data set includes multiple big data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in order of frequency from large to small.

[0089] Furthermore, the data analysis unit is used to analyze the large data set to obtain an optimized word frequency sequence, including:

[0090] The following operations are performed on the big data in the big data collection:

[0091] Extracting information content from the big data to obtain content data, extracting search terms and search results from the content data, and if the search result is a preset no-result value, retaining the search term, otherwise, removing the big data corresponding to the content data;

[0092] The retained search terms are summarized to obtain an optimized term data set, and a recursive operation is performed on the optimized term data set to obtain an optimized term frequency sequence.

[0093] It is understandable that the search term is the word or short text searched in the content data, and the search result is the result when the user searches for the search term in the smart box. For example, if the information content is: User searched for ‘Movie A’, The search result is [none], then the search term is

Movie A

[0094] It should be noted that the purpose of this step is to screen out the film and television resources that the user searches for in the smart box but cannot find, so as to provide data support and target orientation for updating the resource library of the TV box in the cloud. For example, every user in the smart box unit searches for

Movie A

Movie A

Movie A

[0095] Furthermore, performing a recursive operation on the optimized word data set to obtain an optimized word frequency sequence includes:

[0096] Successively extract optimized word data from the optimized word data set, and perform the following operations on the extracted optimized word data:

[0097] Obtain an initial word frequency sequence, where the initial word frequency sequence includes multiple word frequency groups, and the word frequency group includes a target word and a frequency, and the initial value of the frequency is 0;

[0098] If there is a target word corresponding to the optimized word data in the multiple word frequency groups in the initial word frequency sequence, perform an increment operation on the frequency of the word frequency group corresponding to the target word in the initial word frequency sequence to obtain an updated word frequency group, update the initial word frequency sequence with the updated word frequency group, and use the updated initial word frequency sequence as the initial word frequency sequence, and return to the step of successively extracting optimized word data from the optimized word data set;

[0099] Otherwise, construct an initial added word frequency group with the optimized word data as the target word, assign the frequency of the initial added word frequency group to 1 to obtain an added word frequency group, import the added word frequency group into the initial word frequency sequence, use the added word frequency group as the word frequency group, and use the initial word frequency sequence after importing the added word frequency group as the initial word frequency sequence, and return to the step of successively extracting optimized word data from the optimized word data set;

[0100] After confirming that the extraction of the optimized word data in the optimized word data set is completed, perform a sorting operation on the word frequency groups in the initial word frequency sequence in descending order of frequency to obtain an optimized word frequency sequence.

[0101] It is understandable that the initial word frequency sequence is a storage sequence for counting the frequency of occurrence of the optimization word data in the large data set, and the target word is the word stored in the word frequency group. The initial value of the frequency is 0, which means that before the recursive operation is performed on the optimization word data set, the frequency of the target word is 0, and before the recursive operation is performed on the optimization word data set, the target word has no specific value.

[0102] It should be noted that the operation of adding the frequency of the word frequency group corresponding to the target word in the initial word frequency sequence is: adding 1 to the frequency value of the word frequency group corresponding to the target word. The initial added word frequency group is a word frequency group constructed with the optimized word data as the target word and the initial value of the frequency being 0.

[0103] Exemplarily, if the optimized word data set includes three optimized word data and they are [Movie A, Movie A, Movie B] respectively, the word frequency group structure of the initial word frequency sequence is [target word, 0] and there are multiple word frequency groups, the optimized word data [Movie A] is extracted, and at this time there is no target word corresponding to the optimized word data in the multiple word frequency groups in the initial word frequency sequence, then based on the optimized word data [Movie A], an initial added word frequency group is constructed, that is, [Movie A, 0], and the frequency of the initial added word frequency group is assigned to 1 to obtain the added word frequency group [Movie A, 1]. After the added word frequency group is imported into the initial word frequency sequence, the initial word frequency sequence is [Movie A, 1]. Extract the second optimized word data [Movie A]. If the initial word frequency sequence [Movie A, 1] contains the target word corresponding to the optimized word data, then perform an incremental operation on the frequency of [Movie A, 1] to obtain an updated word frequency group [Movie A, 2]. Use the updated word frequency group to update the initial word frequency sequence, that is, use [Movie A, 2] to replace [Movie A, 1] in the initial word frequency sequence, so that the updated initial word frequency sequence is [Movie A, 2].

[0104] S5. Obtain a sliding frequency extraction group, parse the optimized word frequency sequence based on a pre-constructed search trend evaluation method and a sliding frequency extraction group, and obtain a plurality of valuable film and television terms, wherein the sliding frequency extraction group includes a preset number of blank groups.

[0105] Furthermore, the sliding frequency extraction group is obtained, and the optimized word frequency sequence is parsed based on the pre-built search trend evaluation method and the sliding frequency extraction group to obtain multiple valuable film and television entries, including:

[0106] Extract optimized word frequency groups from the optimized word frequency sequence in sequence, and map the extracted optimized word frequency groups to blank groups in the sliding frequency extraction group in sequence, until all blank groups in the sliding frequency extraction group are mapped with optimized word frequency groups, and obtain a term group to be analyzed, wherein the term group to be analyzed includes multiple term frequency groups to be analyzed, the term frequency groups to be analyzed correspond to the optimized word frequency groups one by one, and the multiple term frequency groups to be analyzed are sorted in the order from the first to the last in which the optimized word frequency groups are mapped to the blank groups;

[0107] Determine whether the term group to be analyzed is a pre-constructed value term group. If the term group to be analyzed is a value term group, then based on the term group to be analyzed and the sliding frequency extraction group, slide and intercept the optimized term frequency sequence to obtain an updated term group, and use the updated term group as the term group to be analyzed, and return to the step of determining whether the term group to be analyzed is a pre-constructed value term group until the term group to be analyzed is not a value term group, summarize the value term groups, obtain multiple value term groups, perform duplicate term deduplication operation on the multiple value term groups, and obtain multiple valuable film and television terms.

[0108] It is understandable that the sliding frequency extraction group includes a preset number of blank groups, and the sliding frequency extraction group is a sliding window used for sliding extraction of optimized word frequency sequences.

[0109] For example, the preset number is 3, and the sliding frequency extraction group is [blank group, blank group, blank group]. The optimized word frequency sequence is [movie A, n] [movie b, m] [movie c, l] [movie d, g], where n, m, l and g represent the frequencies of the corresponding target words movie A, movie b, movie c and movie d, respectively. Extract the optimized word frequency group [Movie A, n], map it to the sliding frequency extraction group, and get [Movie A, n, blank group, blank group]; extract the optimized word frequency group [Movie b, m], map it to the sliding frequency extraction group, and get [Movie A, n, Movie b, m, blank group]; extract the optimized word frequency group [Movie c, l], map it to the sliding frequency extraction group, and get [Movie A, n, Movie b, m, Movie c, l], then get the term group to be analyzed [Movie A, n, Movie b, m, Movie c, l], and the term frequency groups to be analyzed are: [Movie A, n], [Movie b, m], [Movie c, l]. The multiple term frequency groups to be analyzed are sorted in the order from the first to the last of the optimized word frequency groups mapped to the blank group, that is, the multiple term frequency groups to be analyzed in the term group to be analyzed are in the same order as in the optimized word frequency sequence.

[0110] Furthermore, the sliding interception of the optimized word frequency sequence is an operation of sequentially intercepting the sliding frequency extraction groups according to the order of the optimized word frequency sequence, and the number of optimized word frequency groups intercepted by each sliding interception is 1.

[0111] Exemplarily, the optimized word frequency sequence is [Movie A, n] [Movie b, m] [Movie c, l] [Movie d, g], [Movie e, f], where f represents the frequency of movie e, and the word frequency groups to be analyzed are: [Movie A, n], [Movie b, m], [Movie c, l], and the updated term group obtained by slidingly intercepting the optimized word frequency sequence is [Movie b, m, Movie c, l, Movie d, g].

[0112] Furthermore, the duplicate term deduplication operation is to remove repeated target words in multiple valuable term groups. It can be understood that according to the previous example, the term group to be analyzed [Movie A, n, Movie b, m, Movie c, l] and the second term group to be analyzed [Movie b, m, Movie c, l, Movie d, g] are obtained, and there are repeated [Movie b, m, Movie c, l] in the term group to be analyzed and the second term group to be analyzed. The duplicate term deduplication operation is performed on the term group to be analyzed and the second term group to be analyzed, and the multiple valuable film and television terms obtained are: Movie A, n, Movie b, m, Movie c, l, Movie d, g.

[0113] Furthermore, the step of judging whether the to-be-analyzed term group is a pre-constructed value term group includes:

[0114] Extract the word frequency groups to be analyzed from the word group to be analyzed in sequence, and perform the following operations on the extracted word frequency groups to be analyzed:

[0115] Obtain search terms corresponding to the word frequency group to be analyzed to obtain a target term, obtain a release date and a data cutoff date based on the target term, obtain a daily search frequency graph based on the release date, the data cutoff date and the target term, construct a daily frequency search volume regression model based on the daily search frequency graph and a pre-constructed frequency difference autoregression model, calculate the search volumes of multiple preset prediction target time points using the daily frequency search volume regression model, obtain multiple predicted search volumes, map the multiple predicted search volumes and the predicted target time points corresponding to each predicted search volume to a pre-constructed blank daily search frequency graph, and obtain a predicted frequency graph, wherein the horizontal axis of the predicted frequency graph is the predicted target time point, and the vertical axis is the predicted search volume, and the predicted target time points correspond to the predicted search volumes one by one;

[0116] Performing a linear fitting operation on the predicted frequency graph to obtain a fitting straight line, obtaining the slope of the fitting straight line, and obtaining a search frequency trend value. If the search frequency trend value is greater than a preset trend threshold, the target term is confirmed as a unit value term;

[0117] Determine whether the number of unit value terms in the term group to be analyzed is greater than a preset value number;

[0118] If the number of unit value terms in the term group to be analyzed is greater than the value number, the term group to be analyzed is confirmed to be a value term group.

[0119] It is understandable that the release date is the premiere date officially released by the target term, the data cutoff date is the last date in the historical period, and the daily search frequency graph is a frequency graph constructed by using the daily search volume of the target term in the smart box unit. The frequency difference autoregression model is a prediction model for daily search volume based on the ARIMA model. The ARIMA model is an autoregression integral moving average model, which can be used for prediction of time series analysis. It processes and predicts time series data by combining autoregression (AR), difference (I) and moving average (MA). Therefore, in an embodiment of the present invention, a frequency difference autoregression model based on the ARIMA model can be used to predict the search volume of multiple preset prediction target time points, and this technology is a prior art and will not be repeated here. The blank day search frequency graph is a frequency graph constructed in the same way as the corresponding daily search frequency graph, but the blank day search frequency graph removes the coordinate values ​​in the daily search frequency graph. The predicted frequency graph is a frequency graph obtained by importing the predicted target time point and the predicted search volume into the blank day search frequency graph one by one.

[0120] Further, in an embodiment of the present invention, a linear fitting operation is performed on the predicted frequency graph, and the fitted straight line is obtained by performing a linear fitting operation using the least square method, and the obtained straight line is the fitted straight line. The search frequency trend value is the slope of the fitted straight line. The trend threshold is the minimum value of the slope of the fitted straight line set manually. When the search frequency trend value is greater than the preset trend threshold, it is considered that the search trend of the target term in the time period corresponding to the multiple predicted target time points meets the manually set requirements. For example, when the search frequency trend value is greater than the preset trend threshold, it is considered that there is still value in introducing the corresponding film and television resources. Therefore, in an embodiment of the present invention, the unit value term is the target term whose search frequency trend value is greater than the preset trend threshold. The value term group is the term group in which the number of unit value terms in the term group to be analyzed is greater than the value number. The value number is a manually set number. When the number of unit value terms in the term group to be analyzed is greater than the value number, it is considered that the film and television resources corresponding to the target words in the term group to be analyzed are all worth introducing.

[0121] Exemplarily, the term group to be analyzed is [Movie A, n, Movie b, m, Movie c, l], and the second term group to be analyzed is [Movie b, m, Movie c, l, Movie d, g]. In the second term group to be analyzed, it is found that the search frequency trend value of [Movie b, m] is greater than the preset trend threshold, which is a unit value term, but Movie and Movie d are not unit value terms, and the value number is artificially set to 2. At this time, the second term group to be analyzed is not a value term group. At this time, the operation of sliding interception of the optimized term frequency sequence based on the term group to be analyzed and the sliding frequency extraction group is stopped, and the second term group to be analyzed is not summarized when summarizing the value term group, and the sliding interception operation is no longer performed on [Movie c, l, Movie d, g] and its lagging optimized term frequency group in the optimized term frequency sequence. This is because the embodiment of the present invention previously arranged the optimized word frequency sequence in descending order of word frequency. When the second term group to be analyzed is not a valuable term group, it means that part or all of the search volume of the lagging optimized term frequency group is not a unit valuable term. Therefore, the sliding interception operation is stopped at this time, saving the amount of calculation while extracting valuable film and television terms as much as possible.

[0122] It is understandable that the daily search frequency graph obtained based on the release date, data cutoff date and target term includes:

[0123] Obtaining a search period based on the release date and the data cutoff date, dividing the search period to obtain a day sequence, and constructing an initial day search frequency coordinate system based on the target term with the day sequence as the horizontal axis and the frequency as the vertical axis;

[0124] The day sequences are extracted from the day sequence in sequence, and the total daily search frequency corresponding to the target term is extracted from the big data set based on the extracted day sequences and the target term, and the total daily search frequency and the corresponding day sequence are mapped to the initial daily search frequency coordinate system. After confirming that all the day sequences in the day sequence have been extracted, the initial daily search frequency graph is obtained.

[0125] Furthermore, the search period is a time interval constructed by the release date and the data cutoff date, and the search period is divided into 24 hours per day with 0:00 as the dividing point. For example, 0:00 on January 1st to 0:00 on January 2nd is a day sequence, and the day sequence is one day. Further, the day sequence is a sequence that sorts the divided search periods in order of date. The direction in which the horizontal axis in the initial day search frequency coordinate system increases is the direction in which the day sequence increases. The total daily search frequency is the total amount of searches for the target term in all smart boxes recorded in the big data set on the date corresponding to the day sequence. For example, day sequence 1 is: January 1st 0:00-January 2nd 0:00, then the total daily search frequency is the total frequency of searching for the target term during the period from January 1st 0:00 to January 2nd 0:00, and mapping the total daily search frequency and the corresponding day sequence to the initial daily search frequency coordinate system means taking the total daily search frequency as the vertical coordinate and the corresponding day sequence as the horizontal coordinate to form coordinate points, and plotting the points in the initial daily search frequency coordinate system.

[0126] Furthermore, the fitting straight line is as follows:

[0127] in, represents the predicted search volume in the fitted line, represents the predicted target time point in the fitting line, Indicates The index of the predicted target time point, Indicates the total number of predicted target time points, Indicates Predict the target time point, Represents the mean of multiple predicted target time points, Indicates Predict the predicted search volume corresponding to the target time point, Represents the mean of the predicted search volumes corresponding to multiple predicted target time points.

[0128] S6. Obtain a time period film and television resource library corresponding to multiple valuable film and television entries, and extract smart boxes from multiple smart boxes in sequence, intercept the initial analysis data set of the smart box based on the preset time period, obtain the time period search data set, parse the time period search data set, and obtain the user search directory, wherein the time period film and television resource library includes multiple film and television resources, and the film and television resources correspond one-to-one to the valuable film and television entries, and the user search directory includes zero, one or more search entries.

[0129] Furthermore, the time period film and television resource library is a collection of film and television resources corresponding to multiple valuable film and television entries. If the time period film and television resource library is provided to the smart box unit, each smart box in the smart box unit can watch the film and television resources on TV. The preset time period is a period of time set manually, which is used to provide users with new film and television prompts based on time. The time period search data set is the initial analysis data set under the preset time period. For example, the initial analysis data set is the past 30 days, and the preset time period is 7 days. Then the time period search data set is a collection of initial analysis data for the past 7 days extracted from the initial analysis data set.

[0130] Specifically, the user search catalog is the terms and frequencies searched by the user in a preset time period. The steps of parsing the time period search data set to obtain the user search catalog are similar to the steps of parsing the big data set to obtain the optimized word frequency sequence, and can achieve the same effect, which will not be repeated here. The search terms are the terms in the user search catalog, which are the terms that the user has searched in the preset time period obtained by parsing the time period search data set.

[0131] S7. If it is confirmed that the film and television resources corresponding to the search terms in the user's search catalog exist in the film and television resource library of the time period, the smart box is used to initiate a new film and television reminder based on the search terms to the user, and the smart box data processing based on edge computing is completed based on the new film and television reminder.

[0132] Exemplarily, the preset time period is 7 days. When it is detected that movie A exists in the user search catalog of the smart box in the past 7 days, and movie A also exists in the film and television resource library of the time period, the smart box is used to prompt the user: [The movie A you are interested in is now available].

[0133] In order to solve the problems described in the background technology, the present invention first obtains an edge computing-smart box analysis system, extracts smart boxes from the smart box units in sequence, and performs the following operations on the extracted smart boxes: based on the smart box receiving the log authorization instruction from the user, the historical search data set of the smart box in the pre-constructed historical period is obtained according to the log authorization instruction, the historical search data set is sent to the edge computing unit, and the edge computing unit that receives the historical search data set performs preprocessing operations on the historical search data set to obtain an initial analysis data set. In the present invention, when the smart box receives the log authorization instruction from the user, the smart box can upload and send the locally retained logs to the edge gateway and other places for processing, so as to provide intelligent services to the user with the user's permission. The log authorization instruction also ensures that the user's usage traces are not leaked. The present invention also uses the historical search data set to filter within the historical period to obtain the historical search data set within the historical period. Aggregate the initial analysis data sets corresponding to multiple smart boxes to obtain a big data set, use the data analysis unit to parse the big data set, and obtain an optimized word frequency sequence, wherein the big data set includes multiple big data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in order from large to small in frequency, obtain a sliding frequency extraction group, and parse the optimized word frequency sequence based on the pre-constructed search trend evaluation method and the sliding frequency extraction group to obtain multiple valuable film and television entries, wherein the sliding frequency extraction group includes a preset number of blank groups. The present invention provides data support and target guidance for filtering out film and television resources that users search for but cannot find in the smart box, thereby updating the TV box resource library in the cloud. The optimized word frequency sequence is evaluated using a sliding frequency extraction group, and a unit-value term is confirmed when the search frequency trend value is greater than a preset trend threshold. When the number of unit-value terms in the sliding frequency extraction group is greater than the preset value number, a value term group is obtained, and the optimized word frequency sequence is slidingly intercepted using the sliding frequency extraction group. When the term group to be analyzed is not a value term group, it means that part or all of the search volume of the lagging optimized word frequency group is not a unit-value term, and its search trend therefore stops the sliding interception operation at this time, saving computational effort while extracting valuable film and television terms as much as possible.Obtain a time period film and television resource library corresponding to multiple valuable film and television terms, and extract smart boxes from multiple smart boxes in sequence, intercept the initial analysis data set of the smart box based on the preset time period, obtain the time period search data set, parse the time period search data set, and obtain the user search directory, wherein the time period film and television resource library includes multiple film and television resources, and the film and television resources correspond to the valuable film and television terms one by one, and the user search directory includes zero, one or more search terms. If it is confirmed that the film and television resources corresponding to the search terms in the user search directory exist in the time period film and television resource library, then use the smart box to initiate a new film and television reminder based on the search terms to the user, and complete the smart box data processing based on edge computing based on the new film and television reminder. After the introduction of the time period film and television resource library, the present invention provides a new reminder for searching for the terms existing in the time period film and television resource library in a short period of time, thereby improving the intelligence of the smart box and the user experience of the user. Therefore, the present invention can use edge computing to parse the search data of the smart box in the past time period to obtain a valuable time period film and television resource library, and provide intelligent new reminders to the user.

[0134] like Figure 2 The figure shows a functional module diagram of a smart box based on edge computing and its data processing system provided by one embodiment of the present invention.

[0135] The edge computing-based smart box and its data processing system 100 of the present invention can be installed in an electronic device. According to the functions implemented, the edge computing-based smart box and its data processing system 100 can include a system acquisition module 101, an initial analysis data set acquisition module 102, a valuable film and television entry acquisition module 103 and a smart box prompt module 104. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0136] The system acquisition module 101 is used to acquire an edge computing-smart box analysis system, wherein the edge computing-smart box analysis system includes: an edge computing unit, a data analysis unit and a smart box unit, wherein the smart box unit includes a plurality of smart boxes;

[0137] The initial analysis data set acquisition module 102 is used to sequentially extract smart boxes from the smart box units, and perform the following operations on the extracted smart boxes: based on the smart box receiving a log authorization instruction from the user, obtain a historical search data set of the smart box in a pre-constructed historical period according to the log authorization instruction, send the historical search data set to the edge computing unit, and perform a preprocessing operation on the historical search data set using the edge computing unit that receives the historical search data set to obtain an initial analysis data set;

[0138] The valuable film and television terms acquisition module 103 is used to aggregate the initial analysis data sets corresponding to the multiple smart boxes to obtain a large data set, and use the data analysis unit to parse the large data set to obtain an optimized word frequency sequence, wherein the large data set includes multiple large data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in order from large to small in frequency, and a sliding frequency extraction group is obtained. The optimized word frequency sequence is parsed based on the pre-constructed search trend evaluation method and the sliding frequency extraction group to obtain multiple valuable film and television terms, wherein the sliding frequency extraction group includes a preset number of blank groups;

[0139] The smart box prompt module 104 is used to obtain a time period film and television resource library corresponding to multiple valuable film and television terms, and extract smart boxes from multiple smart boxes in sequence, intercept the initial analysis data set of the smart box based on a preset time period, obtain a time period search data set, parse the time period search data set, and obtain a user search directory, wherein the time period film and television resource library includes multiple film and television resources, and the film and television resources correspond to valuable film and television terms one by one, and the user search directory includes zero, one or more search terms. If it is confirmed that the film and television resources corresponding to the search terms in the user search directory exist in the time period film and television resource library, the smart box is used to initiate a new film and television prompt based on the search terms to the user, and the smart box data processing based on edge computing is completed based on the new film and television prompt.

[0140] In detail, the modules in the edge computing-based smart box and its data processing system 100 described in the embodiment of the present invention are used in the same manner as described above. Figure 1 The smart box based on edge computing and its data processing method described in the text have the same technical means and can produce the same technical effects, so they will not be repeated here.

[0141] like Figure 3 The figure is a schematic diagram of the structure of an electronic device for implementing a smart box based on edge computing and a data processing method thereof provided by one embodiment of the present invention.

[0142] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a smart box based on edge computing and a data processing method program thereof.

[0143] Among them, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 can also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 also includes an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the smart box based on edge computing and its data processing method program, but also can be used to temporarily store data that has been output or is to be output.

[0144] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (such as a smart box based on edge computing and its data processing method program, etc.), and calls data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0145] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0146] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0147] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management system, so that the power management system can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0148] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0149] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0150] The program of the smart box based on edge computing and the data processing method thereof stored in the memory 11 in the electronic device 1 is a combination of multiple instructions. When running in the processor 10, the following can be achieved:

[0151] Obtain an edge computing-smart box analysis system, wherein the edge computing-smart box analysis system comprises: an edge computing unit, a data analysis unit and a smart box unit, wherein the smart box unit comprises a plurality of smart boxes;

[0152] Extract smart boxes from the smart box unit one by one, and perform the following operations on the extracted smart boxes:

[0153] Based on the smart box receiving the log authorization instruction from the user, obtaining the historical search data set of the smart box in the pre-constructed historical period according to the log authorization instruction, sending the historical search data set to the edge computing unit, and performing a preprocessing operation on the historical search data set by using the edge computing unit that receives the historical search data set to obtain an initial analysis data set;

[0154] Aggregate the initial analysis data sets corresponding to the multiple smart boxes to obtain a big data set, and use the data analysis unit to parse the big data set to obtain an optimized word frequency sequence, wherein the big data set includes multiple big data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in descending order of frequency;

[0155] Obtain a sliding frequency extraction group, parse the optimized word frequency sequence based on a pre-built search trend evaluation method and a sliding frequency extraction group, and obtain a plurality of valuable film and television terms, wherein the sliding frequency extraction group includes a preset number of blank groups;

[0156] Obtain a time period film and television resource library corresponding to a plurality of valuable film and television entries, and extract smart boxes from a plurality of smart boxes in sequence, intercept an initial analysis data set of the smart boxes based on a preset time period, obtain a time period search data set, parse the time period search data set, and obtain a user search directory, wherein the time period film and television resource library includes a plurality of film and television resources, and the film and television resources correspond to the valuable film and television entries one by one, and the user search directory includes zero, one or more search entries;

[0157] If it is confirmed that the film and television resources corresponding to the search terms in the user's search catalog exist in the film and television resource library of the time period, the smart box will be used to initiate a new film and television reminder based on the search terms to the user, and the smart box data processing based on edge computing will be completed based on the new film and television reminder.

[0158] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0159] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or system that can carry the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0160] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0161] Obtain an edge computing-smart box analysis system, wherein the edge computing-smart box analysis system comprises: an edge computing unit, a data analysis unit and a smart box unit, wherein the smart box unit comprises a plurality of smart boxes;

[0162] Extract smart boxes from the smart box unit one by one, and perform the following operations on the extracted smart boxes:

[0163] Based on the smart box receiving the log authorization instruction from the user, obtaining the historical search data set of the smart box in the pre-constructed historical period according to the log authorization instruction, sending the historical search data set to the edge computing unit, and performing a preprocessing operation on the historical search data set by using the edge computing unit that receives the historical search data set to obtain an initial analysis data set;

[0164] Aggregate the initial analysis data sets corresponding to the multiple smart boxes to obtain a big data set, and use the data analysis unit to parse the big data set to obtain an optimized word frequency sequence, wherein the big data set includes multiple big data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in descending order of frequency;

[0165] Obtain a sliding frequency extraction group, parse the optimized word frequency sequence based on a pre-built search trend evaluation method and a sliding frequency extraction group, and obtain a plurality of valuable film and television terms, wherein the sliding frequency extraction group includes a preset number of blank groups;

[0166] Obtain a time period film and television resource library corresponding to a plurality of valuable film and television entries, and extract smart boxes from a plurality of smart boxes in sequence, intercept an initial analysis data set of the smart boxes based on a preset time period, obtain a time period search data set, parse the time period search data set, and obtain a user search directory, wherein the time period film and television resource library includes a plurality of film and television resources, and the film and television resources correspond to the valuable film and television entries one by one, and the user search directory includes zero, one or more search entries;

[0167] If it is confirmed that the film and television resources corresponding to the search terms in the user's search catalog exist in the film and television resource library of the time period, the smart box will be used to initiate a new film and television reminder based on the search terms to the user, and the smart box data processing based on edge computing will be completed based on the new film and television reminder.

[0168] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and actual implementation may have other division methods.

[0169] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0170] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0171] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0172] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A data processing method for a smart box based on edge computing, characterized in that: The method comprises: Obtain an edge computing-smart box analysis system, wherein the edge computing-smart box analysis system comprises: an edge computing unit, a data analysis unit and a smart box unit, wherein the smart box unit comprises a plurality of smart boxes; Extract smart boxes from the smart box unit one by one, and perform the following operations on the extracted smart boxes: Based on the smart box receiving the log authorization instruction from the user, obtaining the historical search data set of the smart box in the pre-constructed historical period according to the log authorization instruction, sending the historical search data set to the edge computing unit, and performing a preprocessing operation on the historical search data set by using the edge computing unit that receives the historical search data set to obtain an initial analysis data set; Aggregate the initial analysis data sets corresponding to the multiple smart boxes to obtain a big data set, and use the data analysis unit to parse the big data set to obtain an optimized word frequency sequence, wherein the big data set includes multiple big data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in descending order of frequency; Obtain a sliding frequency extraction group, parse the optimized word frequency sequence based on a pre-built search trend evaluation method and a sliding frequency extraction group, and obtain a plurality of valuable film and television terms, wherein the sliding frequency extraction group includes a preset number of blank groups; Obtain a time period film and television resource library corresponding to a plurality of valuable film and television entries, and extract smart boxes from a plurality of smart boxes in sequence, intercept an initial analysis data set of the smart boxes based on a preset time period, obtain a time period search data set, parse the time period search data set, and obtain a user search directory, wherein the time period film and television resource library includes a plurality of film and television resources, and the film and television resources correspond to the valuable film and television entries one by one, and the user search directory includes zero, one or more search entries; If it is confirmed that the film and television resources corresponding to the search terms in the user's search catalog exist in the film and television resource library of the time period, the smart box will be used to initiate a new film and television reminder based on the search terms to the user, and the smart box data processing based on edge computing will be completed based on the new film and television reminder.

2. The data processing method of the smart box based on edge computing according to claim 1, characterized in that: The step of obtaining a historical search data set of the smart box in a pre-built historical period according to the log authorization instruction includes: Parsing the log authorization instruction to obtain a historical period and a historical search log, wherein the historical search log includes a plurality of unit search records; Extracting unit search records from the historical search log in sequence, and obtaining the time of the extracted unit search records to obtain the target time. If the target time is within the historical period, retain the extracted unit search records; otherwise, discard the extracted unit search records. The retained unit search records are aggregated to obtain a historical search data set.

3. The data processing method of the smart box based on edge computing as claimed in claim 2, characterized in that: The method of performing a preprocessing operation on the historical search data set by using the edge computing unit that receives the historical search data set to obtain an initial analysis data set includes: The edge computing unit is used to extract historical search data from the historical search data set in sequence, and the following operations are performed on the extracted historical search data: Extract time, information level and information content from historical search data to obtain multiple target information, and perform the following operations on target information in the multiple target information: Determine whether the target information meets the pre-built field constraints. If the target information meets the field constraints, build a key identifier based on the target information, use the key identifier as the key and the target information as the value to build an initial information key-value pair. Otherwise, remove the historical search data corresponding to the target information. After confirming that each of the multiple target information meets the corresponding field constraint condition, the initial information key-value pairs are summarized to obtain initial analysis data corresponding to the historical search data; The initial analysis data are summarized to obtain the initial analysis data set.

4. The data processing method of the smart box based on edge computing as claimed in claim 3, characterized in that: The method of using the data analysis unit to analyze the large data set to obtain an optimized word frequency sequence includes: The following operations are performed on the big data in the big data collection: Extracting information content from the big data to obtain content data, extracting search terms and search results from the content data, and if the search result is a preset no-result value, retaining the search term, otherwise, removing the big data corresponding to the content data; The retained search terms are summarized to obtain an optimized term data set, and a recursive operation is performed on the optimized term data set to obtain an optimized term frequency sequence.

5. The data processing method of the smart box based on edge computing according to claim 4, characterized in that: The recursive operation is performed on the optimized word data set to obtain the optimized word frequency sequence, including: Extract optimized word data from the optimized word data set in sequence, and perform the following operations on the extracted optimized word data: Obtaining an initial word frequency sequence, wherein the initial word frequency sequence includes a plurality of word frequency groups, and the word frequency group includes a target word and a frequency, and the initial value of the frequency is 0; If there is a target word corresponding to the optimized word data in multiple word frequency groups in the initial word frequency sequence, then perform an increment operation on the frequency of the word frequency group corresponding to the target word in the initial word frequency sequence to obtain an updated word frequency group, use the updated word frequency group to update the initial word frequency sequence, and use the updated initial word frequency sequence as the initial word frequency sequence, and return to the step of sequentially extracting the optimized word data from the optimized word data set; Otherwise, construct an initial added word frequency group with the optimized word data as the target word, and assign the frequency of the initial added word frequency group to 1 to obtain the added word frequency group, import the added word frequency group into the initial word frequency sequence, take the added word frequency group as the word frequency group, take the initial word frequency sequence after importing the added word frequency group as the initial word frequency sequence, and return to the step of extracting the optimized word data from the optimized word data set in sequence; After confirming that the optimized word data in the optimized word data set have been extracted, a sorting operation is performed on the word frequency groups in the initial word frequency sequence in order of frequency from large to small to obtain an optimized word frequency sequence.

6. The data processing method of the smart box based on edge computing according to claim 5, characterized in that: The obtaining of the sliding frequency extraction group, parsing the optimized word frequency sequence based on the pre-built search trend evaluation method and the sliding frequency extraction group, and obtaining a plurality of valuable film and television entries, including: Extract optimized word frequency groups from the optimized word frequency sequence in sequence, and map the extracted optimized word frequency groups to blank groups in the sliding frequency extraction group in sequence, until all blank groups in the sliding frequency extraction group are mapped with optimized word frequency groups, and obtain a term group to be analyzed, wherein the term group to be analyzed includes multiple term frequency groups to be analyzed, the term frequency groups to be analyzed correspond to the optimized word frequency groups one by one, and the multiple term frequency groups to be analyzed are sorted in the order from the first to the last in which the optimized word frequency groups are mapped to the blank groups; Determine whether the term group to be analyzed is a pre-constructed value term group. If the term group to be analyzed is a value term group, then based on the term group to be analyzed and the sliding frequency extraction group, slide and intercept the optimized term frequency sequence to obtain an updated term group, and use the updated term group as the term group to be analyzed, and return to the step of determining whether the term group to be analyzed is a pre-constructed value term group until the term group to be analyzed is not a value term group, summarize the value term groups, obtain multiple value term groups, perform duplicate term deduplication operation on the multiple value term groups, and obtain multiple valuable film and television terms.

7. The data processing method of the smart box based on edge computing according to claim 6, characterized in that: The step of judging whether the to-be-analyzed term group is a pre-constructed value term group, if the to-be-analyzed term group is a value term group, includes: Extract the word frequency groups to be analyzed from the word group to be analyzed in sequence, and perform the following operations on the extracted word frequency groups to be analyzed: Obtain search terms corresponding to the word frequency group to be analyzed to obtain a target term, obtain a release date and a data cutoff date based on the target term, obtain a daily search frequency graph based on the release date, the data cutoff date and the target term, construct a daily frequency search volume regression model based on the daily search frequency graph and a pre-constructed frequency difference autoregression model, calculate the search volumes of multiple preset prediction target time points using the daily frequency search volume regression model, obtain multiple predicted search volumes, map the multiple predicted search volumes and the predicted target time points corresponding to each predicted search volume to a pre-constructed blank daily search frequency graph, and obtain a predicted frequency graph, wherein the horizontal axis of the predicted frequency graph is the predicted target time point, and the vertical axis is the predicted search volume, and the predicted target time points correspond to the predicted search volumes one by one; Performing a linear fitting operation on the predicted frequency graph to obtain a fitting straight line, obtaining the slope of the fitting straight line, and obtaining a search frequency trend value. If the search frequency trend value is greater than a preset trend threshold, the target term is confirmed as a unit value term; Determine whether the number of unit value terms in the term group to be analyzed is greater than a preset value number; If the number of unit value terms in the term group to be analyzed is greater than the value number, the term group to be analyzed is confirmed to be a value term group.

8. The data processing method of the smart box based on edge computing according to claim 7, characterized in that: The method of obtaining a daily search frequency graph based on the release date, the data cutoff date and the target term includes: Obtaining a search period based on the release date and the data cutoff date, dividing the search period to obtain a day sequence, and constructing an initial day search frequency coordinate system based on the target term with the day sequence as the horizontal axis and the frequency as the vertical axis; The day sequences are extracted from the day sequence in sequence, and the total daily search frequency corresponding to the target term is extracted from the big data set based on the extracted day sequences and the target term, and the total daily search frequency and the corresponding day sequence are mapped to the initial daily search frequency coordinate system. After confirming that all the day sequences in the day sequence have been extracted, the initial daily search frequency graph is obtained.

9. The data processing method of the smart box based on edge computing according to claim 8, characterized in that: The fitted straight line is as follows: ， in, represents the predicted search volume in the fitted line, represents the predicted target time point in the fitting line, Indicates The index of the predicted target time point, Indicates the total number of predicted target time points, Indicates Predict the target time point, Represents the mean of multiple predicted target time points, Indicates Predict the predicted search volume corresponding to the target time point, Represents the mean of the predicted search volumes corresponding to multiple predicted target time points.

10. A data processing system for a smart box based on edge computing, characterized in that: The system comprises: A system acquisition module, used to acquire an edge computing-smart box analysis system, wherein the edge computing-smart box analysis system includes: an edge computing unit, a data analysis unit and a smart box unit, wherein the smart box unit includes a plurality of smart boxes; The initial analysis data set acquisition module is used to sequentially extract smart boxes from the smart box units, and perform the following operations on the extracted smart boxes: based on the smart box receiving a log authorization instruction from the user, obtain a historical search data set of the smart box in a pre-constructed historical period according to the log authorization instruction, send the historical search data set to the edge computing unit, and perform a preprocessing operation on the historical search data set using the edge computing unit that receives the historical search data set to obtain an initial analysis data set; A valuable film and television entry acquisition module is used to aggregate the initial analysis data sets corresponding to multiple smart boxes to obtain a large data set, and use the data analysis unit to parse the large data set to obtain an optimized word frequency sequence, wherein the large data set includes multiple large data, the optimized word frequency sequence includes multiple optimized word frequency groups, and the multiple optimized word frequency groups are sorted in order from large to small in frequency, and a sliding frequency extraction group is obtained. The optimized word frequency sequence is parsed based on the pre-constructed search trend evaluation method and the sliding frequency extraction group to obtain multiple valuable film and television entries, wherein the sliding frequency extraction group includes a preset number of blank groups; The smart box prompt module is used to obtain a time period film and television resource library corresponding to multiple valuable film and television terms, and extract smart boxes from multiple smart boxes in sequence, intercept the initial analysis data set of the smart box based on the preset time period, obtain the time period search data set, parse the time period search data set, and obtain the user search directory, wherein the time period film and television resource library includes multiple film and television resources, and the film and television resources correspond to the valuable film and television terms one by one, and the user search directory includes zero, one or more search terms. If it is confirmed that the film and television resources corresponding to the search terms in the user search directory exist in the time period film and television resource library, the smart box is used to initiate a new film and television prompt based on the search terms to the user, and the smart box data processing based on edge computing is completed based on the new film and television prompt.

Images