CN113608882B - Information processing method and system based on artificial intelligence and big data and cloud platform - Google Patents
Information processing method and system based on artificial intelligence and big data and cloud platform Download PDFInfo
- Publication number
- CN113608882B CN113608882B CN202111179522.9A CN202111179522A CN113608882B CN 113608882 B CN113608882 B CN 113608882B CN 202111179522 A CN202111179522 A CN 202111179522A CN 113608882 B CN113608882 B CN 113608882B
- Authority
- CN
- China
- Prior art keywords
- thread
- local
- feature extraction
- threads
- extraction processing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5055—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering software capabilities, i.e. software resources associated or available to the machine
Abstract
In the information processing method, system and cloud platform based on artificial intelligence and big data provided by the embodiment of the application, in view of the fact that the AI thread for interest mining is an intelligent thread with set thread performance parameters meeting the set requirements and obtained by centralized positioning of the AI threads with more diversified thread configuration states, therefore, compared with the interest mining by means of the fuzzy selected AI thread, the method can ensure the compatibility degree between the target AI thread and the user behavior big data, thereby avoiding the configuration constraint of the target AI thread when the interest mining is carried out on the big data of the user behavior, this ensures the integrity and accuracy of the resulting session interest mining, including interest topic tags or potential session interests, therefore, reliable decision basis can be provided for subsequent interest demand analysis and business push optimization.
Description
Technical Field
The application relates to the technical field of artificial intelligence and big data, in particular to an information processing method and system based on artificial intelligence and big data and a cloud platform.
Background
In the internet era, Artificial Intelligence (AI) and Big Data (Big Data) are inseparable. With the development of artificial intelligence and big data, the application thereof has penetrated into various aspects of agriculture, industry, commerce, service industry, medical field, etc., becoming an important factor influencing the development of business. Among them, the combination technology of artificial intelligence and big data formation is more and more widely applied in the business service industry. In practical applications, interest mining for user behavior is the main direction. However, the inventor finds, through research and analysis, that it is difficult for related user interest mining technologies to ensure compatibility between a network model and user behavior data, and further difficult to guarantee integrity and accuracy of user interest mining.
Disclosure of Invention
In order to solve the technical problems in the related art, the application provides an information processing method and system based on artificial intelligence and big data and a cloud platform.
In a first aspect, an embodiment of the present application provides an information processing method based on artificial intelligence and big data, which is applied to an information processing cloud platform, and the method includes: determining user behavior big data triggering the session interest mining condition, and transmitting the user behavior big data triggering the session interest mining condition into a target AI thread to obtain a session interest mining condition; the conversation interest mining condition is intended to represent interest topic labels or potential conversation interests mined from the user behavior big data triggering the conversation interest mining condition; the target AI thread is a target local thread of which the set thread performance parameter obtained by positioning in the first pending AI thread meets the set requirement; the first AI to be determined thread comprises a plurality of first local threads; the first local thread satisfies at least one of a first target condition and a second target condition; the first target condition is: the interest mining quality evaluation of the visual saliency expression determined by the first local thread with the difference is inconsistent; the second target condition is: the differentiated first local threads are inconsistent with configuration strategies for visual saliency expression used as different interest mining quality ratings to perform interest mining; the disparity of the first local thread for configuration policy used as a visual saliency expression for different interest mining quality ratings performing interest mining is intended to represent a match list disparity between the interest mining unit and the visual saliency expression for different interest mining quality ratings in the first local thread for disparity.
Under some independently implementable technical solutions, before the passing of the user behavior big data triggering the session interest mining condition into a target AI thread to obtain a session interest mining condition, the method includes: creating the first to-be-determined AI thread based on a thread assembly indication; and positioning a target local thread from the first to-be-determined AI thread, wherein the target local thread is the first local thread with the set thread performance parameter meeting the set requirement, and taking the target local thread as the target AI thread.
Under some independently implementable technical solutions, the first pending AI thread includes: each first local thread comprises a characteristic extraction processing nodes and b interest mining units which are associated with the characteristic extraction processing nodes, and one characteristic extraction processing node in the a characteristic extraction processing nodes is associated with at least one interest mining unit in the b interest mining units; the thread assembly instruction comprises: the number of the node clusters, the number of the interest mining units and the basic thread variable of each feature extraction processing node are different, and the quantitative extraction period of the feature extraction processing nodes corresponding to each interest mining unit is determined; wherein, the quantization extraction period of one feature extraction processing node is as follows: the quantitative comparison result between the data volume of the user behavior big data transmitted into the feature extraction processing node and the data volume of the visual saliency expression determined by the feature extraction processing node is obtained; an interest mining unit corresponds to the quantitative extraction periods of a plurality of characteristic extraction processing nodes and aims to represent that the interest mining unit is associated with different characteristic extraction processing nodes corresponding to the quantitative extraction periods in a first local process with difference; b and a are positive integers.
Under some independently implementable technical solutions, the locating a target local thread from the first to-be-determined AI thread includes: capturing an original local thread from the first AI to be determined thread; the original local thread is a first local thread with a quantitative extraction cycle of the feature extraction processing node meeting a set requirement; the setting requirement is a thread capturing index configured in advance; performing first optimization on each original local thread by means of example user behavior big data obtained from a set cloud storage space to obtain a second undetermined AI thread; and locating the target local thread from the second pending AI thread.
Under some independently implementable technical solutions, the second pending AI thread includes a number of second local threads, and the locating the target local thread from the second pending AI thread includes: capturing the plurality of second local threads, and adjusting and/or fusing to obtain a third AI thread to be determined; the third pending AI threads comprise a plurality of third local threads; transmitting the auxiliary user behavior big data obtained from the set cloud storage space to each third local thread to obtain mining information of each third local thread; determining a set thread performance parameter of each third local thread based on the mining information of each third local thread; and selecting a third local thread with the maximum set thread performance parameter as the target local thread.
Under some independently implementable technical solutions, the transmitting the auxiliary user behavior big data obtained from the set cloud storage space to each third local thread to obtain mining information of each third local thread includes: randomly selecting a third local thread from the third AI threads to be determined as a current local thread; determining a set of thread variables for the current local thread; taking a group of thread variables of the current local thread as a pre-optimization thread variable, and performing second optimization on the current local thread to obtain a group of target thread variables of the current local thread; the super-parameter convergence index in the second optimization is lower than the super-parameter convergence index in the first optimization; replacing the thread variables of the current local thread with the set of target thread variables; transmitting the auxiliary user behavior big data obtained from the set cloud storage space into the current local thread with thread variables improved, and obtaining mining information; and selecting a next third local thread from the third pending AI threads as the current local thread until each third local thread in the third pending AI threads is selected, thereby obtaining mining information of each third local thread.
Under some independently implementable technical solutions, after performing the first optimization on each original local thread by using the example user behavior big data obtained from the set cloud storage space to obtain a second pending AI thread, the method further includes: determining a thread variable of each second local thread; obtaining a transition thread variable of each feature extraction processing node included in each node cluster in the a node clusters based on the thread variable of each second local thread; the determining a set of thread variables for the current local thread includes: and extracting processing nodes based on each feature in the current local thread, selecting corresponding transition thread variables from the obtained transition thread variables to obtain a transition thread variables, and taking the selected a transition thread variables as a group of thread variables of the current local thread.
Under some independently implementable technical solutions, the obtaining a third pending AI thread by capturing and adjusting and/or fusing the number of second local threads includes: randomly selecting a set number of second local threads from the plurality of second local threads as derived threads; adjusting and/or fusing the derived threads to obtain transition threads; the transition threads comprise a plurality of local transition threads; from the plurality of local transition threads, capturing a local transition thread with a quantitative extraction period of the feature extraction processing node meeting the set requirement as a main local thread, or capturing a local transition thread with a quantitative extraction period of the feature extraction processing node meeting the set requirement and an algorithm quantitative index meeting a set judgment index as a main local thread; taking a local thread set formed by the main local thread and the derived thread as a transition pending AI thread, and determining a set thread performance parameter of each transition local thread in the transition pending AI thread; selecting a set number of transition local threads with the maximum set thread performance parameters from the transition pending AI threads; improving the derived threads by means of the selected set number of transition local threads with the maximum set thread performance parameters; and circulating according to the set round number, and taking the finally obtained transition undetermined AI thread as the third undetermined AI thread.
Under some independently implementable technical solutions, the adjustment process includes one or more of the following processing modes: selecting a plurality of local derivative threads from the derivative threads, and determining one or a plurality of random feature extraction processing nodes in each selected local derivative thread as other random feature extraction processing nodes included in a node cluster corresponding to the one or the plurality of feature extraction processing nodes; selecting a plurality of local derivative threads from the derivative threads, and determining the quantitative extraction period of the feature extraction processing node corresponding to one or a plurality of interest mining units in each selected local derivative thread as the quantitative extraction period of other random feature extraction processing nodes in the quantitative extraction periods of the feature extraction processing nodes; determining a first quantization extraction period corresponding to one interest mining unit as a second quantization extraction period, aiming at representing an association architecture of the interest mining unit and a feature extraction processing node, and modifying the association between the interest mining unit and the first feature extraction processing node corresponding to the first quantization extraction period into the association between the interest mining unit and the second feature extraction processing node corresponding to the second quantization extraction period; the fusion treatment comprises one or more than one of the following treatment modes: selecting a plurality of local derivative threads from the derivative threads, and exchanging one or a plurality of feature extraction processing nodes in two selected random local derivative threads; and selecting a plurality of local derivative threads from the derivative threads, and exchanging the quantization extraction cycles corresponding to one or a plurality of interest mining units in the two selected random local derivative threads.
Under some independently implementable technical solutions, the setting requirement includes at least one of: the quantitative extraction period of each feature extraction processing node is not less than a first set judgment index and not more than a second set judgment index; in the b interest mining units, the quantization extraction period of the feature extraction processing node associated with the previous interest mining unit is not less than the quantization extraction period of the feature extraction processing node associated with the next interest mining unit; in the quantitative extraction periods of the feature extraction processing nodes associated with the b interest mining units, the maximum quantitative extraction period is not less than a third set judgment index; in the quantitative extraction periods of the feature extraction processing nodes associated with the b interest mining units, the minimum quantitative extraction period is not greater than the third set judgment index; in the quantization extraction periods of the feature extraction processing nodes associated with the b interest mining units, the maximum quantization extraction period is different from the minimum quantization extraction period; the first set determination index is lower than the third set determination index, which is lower than the second set determination index; and in a plurality of feature extraction processing nodes associated with the b interest mining units, visual saliency expressions output by the feature extraction processing nodes are different.
Under some independently implementable technical solutions, each first local thread includes a feature extraction processing nodes, the visual saliency expression determined by the c-th feature extraction processing node is obtained based on a first visual saliency expression and a second visual saliency expression, wherein the first visual saliency expression is obtained by performing feature extraction on the visual saliency expression determined by the d-th feature extraction processing node by the c-th feature extraction processing node, and the second visual saliency expression is obtained by performing feature extraction on the visual saliency expression determined by the e-th feature extraction processing node by the parallel feature extraction processing node; the interest mining quality evaluation of the visual saliency expression determined by the e-th feature extraction processing node is the same as that of the c-th feature extraction processing node, the extraction period data of the e-th feature extraction processing node is set extraction period data, and the set extraction period pairing condition is met between the extraction period data of the e-th feature extraction processing node and the extraction period data of the f-th feature extraction processing node; a. c and e are positive integers, e is lower than c, and c is not more than a-1, d is 1 less than c, and f is 1 greater than e.
In a second aspect, an embodiment of the present application further provides an information processing system based on artificial intelligence and big data, including an information processing cloud platform and a service user client that communicate with each other; the service user client is used for uploading user behavior big data to the information processing cloud platform; the information processing cloud platform determines user behavior big data triggering a session interest mining condition, and transmits the user behavior big data triggering the session interest mining condition into a target AI thread to obtain a session interest mining condition; further: the conversation interest mining condition is intended to represent interest topic labels or potential conversation interests mined from the user behavior big data triggering the conversation interest mining condition; the target AI thread is a target local thread of which the set thread performance parameter obtained by positioning in the first pending AI thread meets the set requirement; the first AI to be determined thread comprises a plurality of first local threads; the first local thread satisfies at least one of a first target condition and a second target condition; the first target condition is: the interest mining quality evaluation of the visual saliency expression determined by the first local thread with the difference is inconsistent; the second target condition is: the differentiated first local threads are inconsistent with configuration strategies for visual saliency expression used as different interest mining quality ratings to perform interest mining; the disparity of the first local thread for configuration policy used as a visual saliency expression for different interest mining quality ratings performing interest mining is intended to represent a match list disparity between the interest mining unit and the visual saliency expression for different interest mining quality ratings in the first local thread for disparity.
In a third aspect, the present application further provides an information processing cloud platform, including a processor and a memory; the processor is connected with the memory in communication, and the processor is used for reading the computer program from the memory and executing the computer program to realize the method.
According to the information processing method, the system and the cloud platform based on the artificial intelligence and the big data, the big data of the user behaviors triggering the conversation interest mining conditions are determined, the big data of the user behaviors triggering the conversation interest mining conditions are transmitted into the target AI thread to obtain the conversation interest mining conditions, and the conversation interest mining conditions are used for expressing interest subject labels or potential conversation interests mined from the big data of the user behaviors triggering the conversation interest mining conditions; the target AI thread is a target local thread of which the performance parameter of the set thread obtained by positioning in the first pending AI thread meets the set requirement; the first AI thread to be determined comprises a plurality of first local threads, the interest mining quality evaluation of the visual significance expression determined by the first local threads with differences is inconsistent, and/or the configuration strategy of the visual significance expression used for executing different interest mining quality evaluations of the interest mining by the first local threads with differences is inconsistent; wherein the inconsistency of the configuration strategy of the first local thread with difference for the visual saliency expressions used as the different interest mining quality evaluations for performing the interest mining is intended to represent the inconsistency of the matching list between the interest mining unit and the visual saliency expressions of the different interest mining quality evaluations in the first local thread with difference. Based on this, in view of the fact that the AI thread for interest mining is an intelligent thread, the set thread performance parameters of which are obtained by centralized positioning of the AI threads including more diversified thread configuration states meet the set requirements, compared with interest mining by means of the fuzzy selected AI threads, the method can ensure the compatibility degree between the target AI thread and the user behavior big data, thereby avoiding the configuration constraints on the target AI thread when the target AI thread performs interest mining on the user behavior big data, and thus ensuring the integrity and accuracy of the obtained session interest mining conditions, and the session interest mining conditions include interest topic labels or potential session interests, thereby providing reliable decision bases for subsequent interest demand analysis and service push optimization.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
Fig. 1 is a schematic diagram of a hardware structure of an information processing cloud platform according to an embodiment of the present application.
Fig. 2 is a schematic flowchart of an information processing method based on artificial intelligence and big data according to an embodiment of the present application.
Fig. 3 is a communication architecture diagram of an application environment of an information processing method based on artificial intelligence and big data according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
The method provided by the embodiment of the application can be executed in an information processing cloud platform, a computer device or a similar computing device. Taking an example of running on an information processing cloud platform, fig. 1 is a hardware structure block diagram of an information processing cloud platform implementing an information processing method based on artificial intelligence and big data according to an embodiment of the present application. As shown in fig. 1, information handling cloud platform 10 may include one or more (only one shown in fig. 1) processors 102 (processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, and optionally may further include a transmission device 106 for communication functions. It will be understood by those of ordinary skill in the art that the structure shown in fig. 1 is only an illustration, and does not limit the structure of the information processing cloud platform. For example, information handling cloud platform 10 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
The memory 104 can be used for storing computer programs, for example, software programs and modules of application software, such as a computer program corresponding to an artificial intelligence and big data based information processing method in the embodiment of the present application, and the processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the above-mentioned methods. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 104 may further include memory located remotely from processor 102, which may be connected to information handling cloud platform 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used for receiving or transmitting data via a network. The above-described specific example of the network may include a wireless network provided by a communication provider of the information processing cloud platform 10. In one example, the transmission device 106 includes a Network adapter (NIC), which can be connected to other Network devices through a base station so as to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.
Based on this, please refer to fig. 2, fig. 2 is a schematic flowchart of an information processing method based on artificial intelligence and big data according to an embodiment of the present invention, the method is applied to an information processing cloud platform, and further includes the following technical solutions.
And step S21, determining the user behavior big data triggering the conversation interest mining condition.
And step S22, transmitting the user behavior big data triggering the conversation interest mining condition into a target AI thread to obtain the conversation interest mining condition.
In the embodiment of the present application, the session interest mining condition is intended to indicate an interest topic label (interest preference keyword) or a potential session interest (potential requirement or implicit requirement of a user) mined from the big data of the user behavior triggering the session interest mining condition. In addition, the field related to the user behavior big data can be cloud business services such as online payment, teleworking, intelligent education, intelligent medical treatment and block chaining, and the embodiment of the application is not limited. Further, the session interest mining condition may be determined based on an upload period of user behavior big data or a business session scenario, but is not limited thereto.
Furthermore, in order to improve the related problems of the background art, the target AI thread is a target local thread whose set thread performance parameter obtained by positioning in the first pending AI thread meets the set requirement; the first to-be-determined AI thread includes a number of first local threads. An AI thread may be understood as a neural network such as CNN, LSTM, etc., but is not limited thereto.
Based on the above, the first local thread satisfies at least one of a first target condition and a second target condition. In one aspect, the first target condition may be: the interest mining quality evaluation (feature recognition) of the visual saliency expression (feature vector) determined by the first local thread with the difference is inconsistent. On the other hand, the second target condition is: the first local thread with the difference is inconsistent with the configuration policy used as a visual saliency expression for the different interest mining quality ratings that perform the interest mining.
In addition, the disparity of the first local thread with disparity to the configuration strategy used as the visual saliency expression of the different interest mining quality evaluations for performing the interest mining is not consistent, and is intended to mean that the matching list (correspondence) between the interest mining unit and the visual saliency expression of the different interest mining quality evaluations in the disparity first local thread is not consistent.
In some possible embodiments, before the large data of the user behavior triggering the session interest mining condition is transmitted to the target AI thread to obtain the session interest mining situation as described in step S22, the method may further include the following technical solutions described in steps S31 and S32.
Step S31, creating the first to-be-determined AI thread based on the thread assembly instruction.
For example, the thread assembly indication may be a model creation parameter.
And step S32, positioning a target local thread from the first pending AI thread, wherein the target local thread is the first local thread with a set thread performance parameter meeting a set requirement, and taking the target local thread as the target AI thread.
For example, the set thread performance parameter may be mining accuracy.
On the basis of the above, the first to-be-determined AI thread may include the following: each first local thread comprises a feature extraction processing nodes and b interest mining units which are associated with the feature extraction processing nodes, and one feature extraction processing node in the feature extraction processing nodes is associated with at least one interest mining unit in the b interest mining units. For example, the feature extraction processing node may be a convolution kernel, and the interest mining unit may be an interest miner or a classifier.
Further, the thread assembly instruction includes: the number of the node clusters, the number of the interest mining units and the basic thread variable of each feature extraction processing node are the same, and the quantitative extraction period of the feature extraction processing nodes corresponding to each interest mining unit is obtained.
In some possible embodiments, the quantization extraction period of one feature extraction processing node is: and the quantitative comparison result between the data volume of the user behavior big data transmitted into the feature extraction processing node and the data volume of the visual saliency expression determined by the feature extraction processing node. The quantitative comparison result can be understood as a ratio. The unit of the data amount may be MB or GB.
Furthermore, one interest mining unit corresponds to the quantitative extraction periods of a plurality of feature extraction processing nodes and is used for indicating that the interest mining unit is associated with different feature extraction processing nodes corresponding to the quantitative extraction periods in a first local process with difference; b and a are positive integers.
For some possible examples, the locating the target local thread from the first to-be-determined AI thread described in step S32 may include the technical solutions described in step S321 to step S323.
Step S321, capturing an original local thread from the first to-be-determined AI thread; the original local thread is a first local thread with a quantitative extraction cycle of the feature extraction processing node meeting a set requirement; the setting requirement is a thread capture index configured in advance. For example, a thread capture metric may be understood as a model screening condition.
Step S322, performing first optimization on each original local thread by means of the big data of the example user behaviors obtained from the set cloud storage space, and obtaining a second pending AI thread.
And step S323, positioning the target local thread from the second pending AI thread.
Illustratively, the second pending AI thread may include a number of second local threads. Based on this, the locating the target local thread from the second pending AI thread described in step S323 may include the following technical solutions described in steps S3231 to S3234.
Step S3231, capturing the plurality of second local threads, and adjusting and/or fusing to obtain a third AI thread to be determined; the third pending AI thread includes a number of third local threads.
In some further embodiments, the capturing, adjusting and/or fusing the number of second local threads to obtain the third pending AI thread described in step S3231 may include the following: randomly selecting a set number of second local threads from the plurality of second local threads as derived threads; adjusting and/or fusing the derived threads to obtain transition threads; the transition threads comprise a plurality of local transition threads; from the plurality of local transition threads, capturing a local transition thread with a quantitative extraction period of the feature extraction processing node meeting the set requirement as a main local thread, or capturing a local transition thread with a quantitative extraction period of the feature extraction processing node meeting the set requirement and an algorithm quantitative index meeting a set judgment index as a main local thread; taking a local thread set formed by the main local thread and the derived thread as a transition pending AI thread, and determining a set thread performance parameter of each transition local thread in the transition pending AI thread; selecting a set number of transition local threads with the maximum set thread performance parameters from the transition pending AI threads; improving the derived threads by means of the selected set number of transition local threads with the maximum set thread performance parameters; and circulating according to the set round number, and taking the finally obtained transition undetermined AI thread as the third undetermined AI thread.
In this way, the performance stability of the third pending AI thread can be ensured.
Step S3232, transmitting the auxiliary user behavior big data obtained from the set cloud storage space to each third local thread, and obtaining mining information of each third local thread. For example, auxiliary user behavior big data may be understood as test data.
For some possible embodiments, the transmitting the auxiliary user behavior big data obtained from the set cloud storage space to each third local thread to obtain mining information of each third local thread described in step S3232 may include the following: randomly selecting a third local thread from the third AI threads to be determined as a current local thread; determining a set of thread variables for the current local thread; taking a group of thread variables of the current local thread as a pre-optimization thread variable, and performing second optimization on the current local thread to obtain a group of target thread variables of the current local thread; the super-parameter convergence index in the second optimization is lower than the super-parameter convergence index in the first optimization; replacing the thread variables of the current local thread with the set of target thread variables; transmitting the auxiliary user behavior big data obtained from the set cloud storage space into the current local thread with thread variables improved, and obtaining mining information; and selecting a next third local thread from the third pending AI threads as the current local thread until each third local thread in the third pending AI threads is selected, thereby obtaining mining information of each third local thread.
By the design, the mining information can be completely and accurately determined.
Further, the adjustment processing includes one or more of the following processing modes: selecting a plurality of local derivative threads from the derivative threads, and determining one or a plurality of random feature extraction processing nodes in each selected local derivative thread as other random feature extraction processing nodes included in a node cluster corresponding to the one or the plurality of feature extraction processing nodes; selecting a plurality of local derivative threads from the derivative threads, and determining the quantitative extraction period of the feature extraction processing node corresponding to one or a plurality of interest mining units in each selected local derivative thread as the quantitative extraction period of other random feature extraction processing nodes in the quantitative extraction periods of the feature extraction processing nodes; the method comprises the steps that a first quantization extraction period corresponding to one interest mining unit is determined as a second quantization extraction period, the purpose is to represent an association framework of the interest mining unit and a feature extraction processing node, and the association of the interest mining unit and the first feature extraction processing node corresponding to the first quantization extraction period is corrected to be the association of the interest mining unit and the second feature extraction processing node corresponding to the second quantization extraction period.
In addition, the fusion processing includes one or more of the following processing modes: selecting a plurality of local derivative threads from the derivative threads, and exchanging one or a plurality of feature extraction processing nodes in two selected random local derivative threads; and selecting a plurality of local derivative threads from the derivative threads, and exchanging the quantization extraction cycles corresponding to one or a plurality of interest mining units in the two selected random local derivative threads.
Step S3233 determines the set thread performance parameter for each third local thread based on the mining information for each third local thread.
And step S3234, selecting the third local thread with the maximum set thread performance parameter as the target local thread.
By the design, the precision and the scene adaptation capability of the target local thread can be ensured.
For some independently implementable technical solutions, after performing the first optimization on each original local thread by using the example user behavior big data obtained from the set cloud storage space to obtain the second pending AI thread, the method may further include: determining a thread variable of each second local thread; and obtaining a transition thread variable of each feature extraction processing node included in each node cluster in the a node clusters based on the thread variable of each second local thread. Based on this, determining a set of thread variables for the current local thread includes: and extracting processing nodes based on each feature in the current local thread, selecting corresponding transition thread variables from the obtained transition thread variables to obtain a transition thread variables, and taking the selected a transition thread variables as a group of thread variables of the current local thread.
Illustratively, the setting requirement includes at least one of the following 7 requirements.
(1) The quantization extraction period of each feature extraction processing node is not less than the first set judgment index and not more than the second set judgment index.
(2) In the b interest mining units, the quantization extraction period of the feature extraction processing node associated with the previous interest mining unit is not less than the quantization extraction period of the feature extraction processing node associated with the next interest mining unit.
(3) And in the quantitative extraction periods of the feature extraction processing nodes associated with the b interest mining units, the maximum quantitative extraction period is not less than a third set judgment index.
(4) And in the quantization extraction periods of the feature extraction processing nodes associated with the b interest mining units, the minimum quantization extraction period is not greater than the third set judgment index.
(5) And in the quantization extraction periods of the feature extraction processing nodes associated with the b interest mining units, the maximum quantization extraction period is different from the minimum quantization extraction period.
(6) The first set determination index is lower than the third set determination index, which is lower than the second set determination index.
(7) And in a plurality of feature extraction processing nodes associated with the b interest mining units, visual saliency expressions output by the feature extraction processing nodes are different.
In some possible embodiments, each first local thread includes a feature extraction processing nodes, the visual saliency expression determined by the c-th feature extraction processing node is obtained based on a first visual saliency expression obtained by feature extraction of the visual saliency expression determined by the d-th feature extraction processing node by the c-th feature extraction processing node, and a second visual saliency expression obtained by feature extraction of the visual saliency expression determined by the e-th feature extraction processing node by the parallel feature extraction processing node.
Furthermore, the interest mining quality evaluation of the visual saliency expression determined by the e-th feature extraction processing node is the same as that of the c-th feature extraction processing node, the extraction period data of the e-th feature extraction processing node is set extraction period data, and the set extraction period pairing condition is met between the extraction period data of the e-th feature extraction processing node and the extraction period data of the f-th feature extraction processing node; a. c and e are positive integers, e is lower than c, and c is not more than a-1, d is 1 less than c, and f is 1 greater than e.
By means of the design, the information processing method, the system and the cloud platform based on the artificial intelligence and the big data, which are provided by the embodiment of the application, determine the big data of the user behaviors triggering the session interest mining conditions, and transmit the big data of the user behaviors triggering the session interest mining conditions into the target AI thread to obtain the session interest mining conditions, wherein the session interest mining conditions are intended to represent interest topic tags or potential session interests mined from the big data of the user behaviors triggering the session interest mining conditions; the target AI thread is a target local thread of which the performance parameter of the set thread obtained by positioning in the first pending AI thread meets the set requirement; the first AI thread to be determined comprises a plurality of first local threads, the interest mining quality evaluation of the visual significance expression determined by the first local threads with differences is inconsistent, and/or the configuration strategy of the visual significance expression used for executing different interest mining quality evaluations of the interest mining by the first local threads with differences is inconsistent; wherein the inconsistency of the configuration strategy of the first local thread with difference for the visual saliency expressions used as the different interest mining quality evaluations for performing the interest mining is intended to represent the inconsistency of the matching list between the interest mining unit and the visual saliency expressions of the different interest mining quality evaluations in the first local thread with difference. Based on this, in view of the fact that the AI thread for interest mining is an intelligent thread, the set thread performance parameters of which are obtained by centralized positioning of the AI threads including more diversified thread configuration states meet the set requirements, compared with interest mining by means of the fuzzy selected AI threads, the method can ensure the compatibility degree between the target AI thread and the user behavior big data, thereby avoiding the configuration constraints on the target AI thread when the target AI thread performs interest mining on the user behavior big data, and thus ensuring the integrity and accuracy of the obtained session interest mining conditions, and the session interest mining conditions include interest topic labels or potential session interests, thereby providing reliable decision bases for subsequent interest demand analysis and service push optimization.
Based on the same or similar inventive concepts, an architecture schematic diagram of an application environment 30 of an information processing method based on artificial intelligence and big data is also provided, the application environment comprises an information processing cloud platform 10 and a service user client 20 which are communicated with each other, and the service user client 20 is used for uploading user behavior big data to the information processing cloud platform; the information processing cloud platform 10 determines the user behavior big data triggering the session interest mining condition, and transmits the user behavior big data triggering the session interest mining condition to the target AI thread to obtain the session interest mining condition.
Further, a readable storage medium is provided, on which a program is stored, which when executed by a processor implements the method described above.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus and method embodiments described above are illustrative only, as the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a media service server 10, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (10)
1. An information processing method based on artificial intelligence and big data is applied to an information processing cloud platform, and the method comprises the following steps:
determining user behavior big data triggering the session interest mining condition, and transmitting the user behavior big data triggering the session interest mining condition into a target AI thread to obtain a session interest mining condition;
the conversation interest mining condition is intended to represent interest topic labels or potential conversation interests mined from the user behavior big data triggering the conversation interest mining condition; the target AI thread is a target local thread of which the set thread performance parameter obtained by positioning in the first pending AI thread meets the set requirement; the first AI to be determined thread comprises a plurality of first local threads;
the first local thread satisfies at least one of a first target condition and a second target condition; the first target condition is: the interest mining quality evaluation of the visual saliency expression determined by the first local thread with the difference is inconsistent; the second target condition is: the differentiated first local threads are inconsistent with configuration strategies for visual saliency expression used as different interest mining quality ratings to perform interest mining;
the disparity of the first local thread for configuration policy used as a visual saliency expression for different interest mining quality ratings performing interest mining is intended to represent a match list disparity between the interest mining unit and the visual saliency expression for different interest mining quality ratings in the first local thread for disparity.
2. The method according to claim 1, wherein each first local thread comprises a feature extraction processing nodes, the visual saliency expression determined by the c-th feature extraction processing node is obtained based on a first visual saliency expression and a second visual saliency expression, wherein the first visual saliency expression is obtained by performing feature extraction on the visual saliency expression determined by the d-th feature extraction processing node by the c-th feature extraction processing node, and the second visual saliency expression is obtained by performing feature extraction on the visual saliency expression determined by the e-th feature extraction processing node by the parallel feature extraction processing node; the parallel feature extraction processing nodes are the feature extraction processing nodes with synchronous processing characteristics determined according to the convolution kernel in all the feature extraction processing nodes of the target AI thread;
the interest mining quality evaluation of the visual saliency expression determined by the e-th feature extraction processing node is the same as that of the c-th feature extraction processing node, the extraction period data of the e-th feature extraction processing node is set extraction period data, and the set extraction period pairing condition is met between the extraction period data of the e-th feature extraction processing node and the extraction period data of the f-th feature extraction processing node; a. c and e are positive integers, e is lower than c, and c is not more than a-1, d is 1 less than c, and f is 1 greater than e.
3. The artificial intelligence and big data based information processing method of claim 1, wherein before said passing the big data of user behavior triggering the session interest mining condition into the target AI thread for the session interest mining condition, the method further comprises:
creating the first to-be-determined AI thread based on a thread assembly indication;
and positioning a target local thread from the first to-be-determined AI thread, wherein the target local thread is the first local thread with the set thread performance parameter meeting the set requirement, and taking the target local thread as the target AI thread.
4. The artificial intelligence and big data based information processing method according to claim 3, wherein the first to-be-determined AI thread includes: each first local thread comprises a characteristic extraction processing nodes and b interest mining units which are associated with the characteristic extraction processing nodes, and one characteristic extraction processing node in the a characteristic extraction processing nodes is associated with at least one interest mining unit in the b interest mining units; the thread assembly instruction comprises: the number of the node clusters, the number of the interest mining units and the basic thread variable of each feature extraction processing node are different, and the quantitative extraction period of the feature extraction processing nodes corresponding to each interest mining unit is determined; wherein, the quantization extraction period of one feature extraction processing node is as follows: the quantitative comparison result between the data volume of the user behavior big data transmitted into the feature extraction processing node and the data volume of the visual saliency expression determined by the feature extraction processing node is obtained; an interest mining unit corresponds to the quantitative extraction periods of a plurality of characteristic extraction processing nodes and aims to represent that the interest mining unit is associated with different characteristic extraction processing nodes corresponding to the quantitative extraction periods in a first local process with difference; b and a are positive integers.
5. The method of claim 4, wherein locating a target local thread from the first pending AI thread comprises:
capturing an original local thread from the first AI to be determined thread; the original local thread is a first local thread with a quantitative extraction cycle of the feature extraction processing node meeting a set requirement; the setting requirement is a thread capturing index configured in advance;
performing first optimization on each original local thread by means of example user behavior big data obtained from a set cloud storage space to obtain a second undetermined AI thread;
and locating the target local thread from the second pending AI thread.
6. The method of claim 5, wherein the second pending AI thread comprises a number of second local threads, the locating the target local thread from the second pending AI thread comprising:
capturing the plurality of second local threads, and adjusting and/or fusing to obtain a third AI thread to be determined; the third pending AI threads comprise a plurality of third local threads;
transmitting the auxiliary user behavior big data obtained from the set cloud storage space to each third local thread to obtain mining information of each third local thread;
determining a set thread performance parameter of each third local thread based on the mining information of each third local thread;
selecting a third local thread with the maximum set thread performance parameter as the target local thread;
the step of transmitting the auxiliary user behavior big data obtained from the set cloud storage space to each third local thread to obtain mining information of each third local thread includes:
randomly selecting a third local thread from the third AI threads to be determined as a current local thread;
determining a set of thread variables for the current local thread;
taking a group of thread variables of the current local thread as a pre-optimization thread variable, and performing second optimization on the current local thread to obtain a group of target thread variables of the current local thread; the super-parameter convergence index in the second optimization is lower than the super-parameter convergence index in the first optimization;
replacing the thread variables of the current local thread with the set of target thread variables;
transmitting the auxiliary user behavior big data obtained from the set cloud storage space into the current local thread with thread variables improved, and obtaining mining information;
and selecting a later third local thread from the third pending AI threads as the current local thread until each third local thread in the third pending AI threads is selected, thereby obtaining mining information of each third local thread.
7. The method of claim 6, wherein after the first optimizing each original local thread by the example user behavior big data obtained from the set cloud storage space, obtaining a second pending AI thread, the method further comprises: determining a thread variable of each second local thread; obtaining a transition thread variable of each feature extraction processing node included in each node cluster in the a node clusters based on the thread variable of each second local thread;
the determining a set of thread variables for the current local thread includes: extracting processing nodes based on each feature in the current local thread, selecting corresponding transition thread variables from the obtained transition thread variables to obtain a transition thread variables, and taking the selected a transition thread variables as a group of thread variables of the current local thread;
the obtaining a third to-be-determined AI thread by capturing, adjusting and/or fusing the plurality of second local threads includes:
randomly selecting a set number of second local threads from the plurality of second local threads as derived threads; adjusting and/or fusing the derived threads to obtain transition threads; the transition threads comprise a plurality of local transition threads;
from the plurality of local transition threads, capturing a local transition thread with a quantitative extraction period of the feature extraction processing node meeting the set requirement as a main local thread, or capturing a local transition thread with a quantitative extraction period of the feature extraction processing node meeting the set requirement and an algorithm quantitative index meeting a set judgment index as a main local thread;
taking a local thread set formed by the main local thread and the derived thread as a transition pending AI thread, and determining a set thread performance parameter of each transition local thread in the transition pending AI thread;
selecting a set number of transition local threads with the maximum set thread performance parameters from the transition pending AI threads; improving the derived threads by means of the selected set number of transition local threads with the maximum set thread performance parameters;
circulating according to the set round number, and taking the finally obtained transitional undetermined AI thread as the third undetermined AI thread;
wherein, the adjustment processing comprises one or more than one of the following processing modes: selecting a plurality of local derivative threads from the derivative threads, and determining one or a plurality of random feature extraction processing nodes in each selected local derivative thread as other random feature extraction processing nodes included in a node cluster corresponding to the one or the plurality of feature extraction processing nodes; selecting a plurality of local derivative threads from the derivative threads, and determining the quantitative extraction period of the feature extraction processing node corresponding to one or a plurality of interest mining units in each selected local derivative thread as the quantitative extraction period of other random feature extraction processing nodes in the quantitative extraction periods of the feature extraction processing nodes; determining a first quantization extraction period corresponding to one interest mining unit as a second quantization extraction period, aiming at representing an association architecture of the interest mining unit and a feature extraction processing node, and modifying the association between the interest mining unit and the first feature extraction processing node corresponding to the first quantization extraction period into the association between the interest mining unit and the second feature extraction processing node corresponding to the second quantization extraction period;
the fusion treatment comprises one or more than one of the following treatment modes: selecting a plurality of local derivative threads from the derivative threads, and exchanging one or a plurality of feature extraction processing nodes in two selected random local derivative threads; and selecting a plurality of local derivative threads from the derivative threads, and exchanging the quantization extraction cycles corresponding to one or a plurality of interest mining units in the two selected random local derivative threads.
8. The method of claim 5, wherein the set requirements include at least one of: the quantitative extraction period of each feature extraction processing node is not less than a first set judgment index and not more than a second set judgment index; in the b interest mining units, the quantization extraction period of the feature extraction processing node associated with the previous interest mining unit is not less than the quantization extraction period of the feature extraction processing node associated with the next interest mining unit; in the quantitative extraction periods of the feature extraction processing nodes associated with the b interest mining units, the maximum quantitative extraction period is not less than a third set judgment index; in the quantitative extraction periods of the feature extraction processing nodes associated with the b interest mining units, the minimum quantitative extraction period is not greater than the third set judgment index; in the quantization extraction periods of the feature extraction processing nodes associated with the b interest mining units, the maximum quantization extraction period is different from the minimum quantization extraction period; the first set determination index is lower than the third set determination index, which is lower than the second set determination index; and in a plurality of feature extraction processing nodes associated with the b interest mining units, visual saliency expressions output by the feature extraction processing nodes are different.
9. An information processing system based on artificial intelligence and big data is characterized by comprising an information processing cloud platform and a service user client which are communicated with each other;
the service user client is used for uploading user behavior big data to the information processing cloud platform;
the information processing cloud platform determines user behavior big data triggering a session interest mining condition, and transmits the user behavior big data triggering the session interest mining condition into a target AI thread to obtain a session interest mining condition;
wherein: the conversation interest mining condition is intended to represent interest topic labels or potential conversation interests mined from the user behavior big data triggering the conversation interest mining condition; the target AI thread is a target local thread of which the set thread performance parameter obtained by positioning in the first pending AI thread meets the set requirement; the first AI to be determined thread comprises a plurality of first local threads; the first local thread satisfies at least one of a first target condition and a second target condition; the first target condition is: the interest mining quality evaluation of the visual saliency expression determined by the first local thread with the difference is inconsistent; the second target condition is: the differentiated first local threads are inconsistent with configuration strategies for visual saliency expression used as different interest mining quality ratings to perform interest mining; the disparity of the first local thread for configuration policy used as a visual saliency expression for different interest mining quality ratings performing interest mining is intended to represent a match list disparity between the interest mining unit and the visual saliency expression for different interest mining quality ratings in the first local thread for disparity.
10. An information processing cloud platform, comprising a processor and a memory; the processor is connected in communication with the memory, and the processor is configured to read the computer program from the memory and execute the computer program to implement the method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111179522.9A CN113608882B (en) | 2021-10-11 | 2021-10-11 | Information processing method and system based on artificial intelligence and big data and cloud platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111179522.9A CN113608882B (en) | 2021-10-11 | 2021-10-11 | Information processing method and system based on artificial intelligence and big data and cloud platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113608882A CN113608882A (en) | 2021-11-05 |
| CN113608882B true CN113608882B (en) | 2022-01-28 |
Family
ID=78343512
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111179522.9A Active CN113608882B (en) | 2021-10-11 | 2021-10-11 | Information processing method and system based on artificial intelligence and big data and cloud platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113608882B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113643007B (en) * | 2021-10-14 | 2022-02-08 | 杭银消费金融股份有限公司 | Visualization-based research and development project work order processing method and system |
| CN114218568B (en) * | 2021-12-10 | 2022-08-23 | 厦门吉快科技有限公司 | Big data attack processing method and system applied to cloud service |
| CN114322270A (en) * | 2022-03-03 | 2022-04-12 | 广州海洁尔医疗设备有限公司 | Air purification detection control system and method |
| CN114299683B (en) * | 2022-03-09 | 2022-08-05 | 广东铭昌消防机电工程有限公司 | Intelligent monitoring method, system and device for electrical fire and cloud platform |
| CN114896306B (en) * | 2022-07-14 | 2022-09-16 | 泰山学院 | Data mining method and system based on artificial intelligence model |
| CN116755869B (en) * | 2023-08-23 | 2023-10-27 | 北京浩然五洲软件技术有限公司 | Financial business part processing method, system and medium |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101359333B (en) * | 2008-05-23 | 2010-06-16 | 中国科学院软件研究所 | Parallel data processing method based on latent dirichlet allocation model |
| US8185482B2 (en) * | 2009-03-30 | 2012-05-22 | Microsoft Corporation | Modeling semantic and structure of threaded discussions |
| CN104036451B (en) * | 2014-06-20 | 2018-12-11 | 深圳市腾讯计算机系统有限公司 | Model method for parallel processing and device based on multi-graphics processor |
| US9424092B2 (en) * | 2014-09-26 | 2016-08-23 | Microsoft Technology Licensing, Llc | Heterogeneous thread scheduling |
| CN106408343A (en) * | 2016-09-23 | 2017-02-15 | 广州李子网络科技有限公司 | Modeling method and device for user behavior analysis and prediction based on BP neural network |
| US11360808B2 (en) * | 2017-04-09 | 2022-06-14 | Intel Corporation | Efficient thread group scheduling |
| CN110717097A (en) * | 2019-09-06 | 2020-01-21 | 中国平安财产保险股份有限公司 | Service recommendation method and device, computer equipment and storage medium |
| CN112463205B (en) * | 2020-11-24 | 2021-07-30 | 青岛日日顺乐信云科技有限公司 | AI and big data based application program management method and artificial intelligence server |
| CN113032522A (en) * | 2021-03-19 | 2021-06-25 | 聊城大学 | Text parallel data mining system and method |
| CN113313479A (en) * | 2021-06-25 | 2021-08-27 | 东莞市爱学爱玩教育科技有限公司 | Payment service big data processing method and system based on artificial intelligence |
-
2021
- 2021-10-11 CN CN202111179522.9A patent/CN113608882B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113608882A (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113608882B (en) | Information processing method and system based on artificial intelligence and big data and cloud platform | |
| CN110147722A (en) | A kind of method for processing video frequency, video process apparatus and terminal device | |
| CN111782491B (en) | Disk failure prediction method, device, equipment and storage medium | |
| CN115858796A (en) | Fault knowledge graph construction method and device | |
| WO2023169274A1 (en) | Data processing method and device, and storage medium and processor | |
| CN112860676B (en) | Data cleaning method applied to big data mining and business analysis and cloud server | |
| WO2022002068A1 (en) | Data processing method, system and device and storage medium | |
| CN112352412A (en) | Network traffic processing method and device, storage medium and computer equipment | |
| CN114221991B (en) | Session recommendation feedback processing method based on big data and deep learning service system | |
| US20240095529A1 (en) | Neural Network Optimization Method and Apparatus | |
| CN111817935B (en) | Internet intelligent home data processing method and system | |
| CN115563069B (en) | Data sharing processing method and system based on artificial intelligence and cloud platform | |
| CN112541556A (en) | Model construction optimization method, device, medium, and computer program product | |
| CN112861894A (en) | Data stream classification method, device and system | |
| CN112926952A (en) | Cloud computing-combined big data office business processing method and big data server | |
| CN114661886A (en) | Online business topic processing method and server applied to big data intelligent analysis | |
| CN112633334A (en) | Modeling method based on satellite measurement, operation and control resource planning and scheduling | |
| CN112925831A (en) | Big data mining method and big data mining service system based on cloud computing service | |
| Zhang et al. | Near-optimal distributed detection in balanced binary relay trees | |
| CN115185780B (en) | Data acquisition method and system based on industrial Internet | |
| CN112055076A (en) | Multifunctional intelligent monitoring method and device based on Internet and server | |
| CN112990425A (en) | Automatic classification method of 5G network slices, device thereof, electronic equipment and computer storage medium | |
| CN115237503B (en) | VR-based ecological model building method and system | |
| CN116192629B (en) | Differential data processing method and system based on edge calculation | |
| CN116566846B (en) | Model management method and system, shared node and network node |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 510000 room h-1408, building 28, No.61 Dalingshan Road, Tianhe District, Guangzhou City, Guangdong Province Patentee after: GUANGZHOU ZMOFFICE TECHNOLOGY CO.,LTD. Address before: 510000 room h-1408, building 28, No.61 Dalingshan Road, Tianhe District, Guangzhou City, Guangdong Province Patentee before: Guangzhou Purple Wheat Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220921 Address after: Room 239, Room 406, No. 1 Yichuang Street, Huangpu District, Guangzhou City, Guangdong Province (Sino-Singapore Guangzhou Knowledge City) 510000 Patentee after: Guangdong Wheat Information Technology Co.,Ltd. Address before: 510000 room h-1408, building 28, No.61 Dalingshan Road, Tianhe District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU ZMOFFICE TECHNOLOGY CO.,LTD. |