User demand analysis method and system applied to solar cells

Abstract

The application relates to a user demand analysis method and system applied to solar cells. Globalization arrangement of feedback text word vectors of different attention indexes can comprehensively analyze the contribution of the feedback text word vectors of different attention indexes to demand keyword marking and the mutual influence between the feedback text word vectors of different attention indexes, so that the feedback text global word vector obtained through the globalization arrangement can accurately and reliably reflect the commonality comparison result of user demand matters in the use feedback of two groups of solar cells, and further guarantee the precision and reliability of demand keyword marking combined with the feedback text global word vector.

Description

Technical Field

[0001] This invention relates to the field of information processing technology, and more specifically, to a user demand analysis method and system for solar cells. Background Technology

[0002] Solar cells are used to directly convert sunlight into electricity, offering advantages such as a continuous supply of solar energy, clean energy, and environmental friendliness. With the development of solar cells, their applications have expanded beyond large-scale factory production to include everyday residential use. However, there are still some areas for improvement in the application of solar cells in residential settings. For example, accurately and reliably analyzing user needs for solar cells in practical applications is a widely discussed issue. Summary of the Invention

[0003] In a first aspect, embodiments of the present invention provide a user demand analysis method for solar cells, applied to a user demand analysis system, the method comprising:

[0004] The system sequentially obtains the first solar cell usage feedback to be marked with demand keywords and the second solar cell usage feedback after the demand keywords have been marked, along with the feedback text word vectors for several key indicators. The feedback text word vectors for these key indicators include at least: demand item sentiment word vectors and usage scenario experience word vectors. The demand item sentiment word vectors reflect the user demand sentiment field of the user demand item in the solar cell usage feedback, and the usage scenario experience word vectors reflect the scenario-based usage experience field of the user demand item in the solar cell usage feedback.

[0005] The word vectors of the feedback texts on several key indicators of the first solar cell usage feedback and the word vectors of the feedback texts on several key indicators of the second solar cell usage feedback are globally organized to obtain global word vectors of feedback texts that reflect the commonality of details of user needs in the first solar cell usage feedback and user needs in the second solar cell usage feedback.

[0006] The global word vector of the feedback text is used to determine whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords; wherein, solar cell usage feedback that matches the same demand keywords is solar cell usage feedback for the same user demand.

[0007] In some optional embodiments, the step of globally organizing the feedback text word vectors of the first solar cell usage feedback on several key indicators and the feedback text word vectors of the second solar cell usage feedback on several key indicators to obtain global feedback text word vectors reflecting the commonality values ​​of user needs in the first solar cell usage feedback and user needs details in the second solar cell usage feedback includes:

[0008] For one of the several attention indicators, a common description array of the details of the attention indicator is determined by combining the feedback text word vector of the first solar cell usage feedback on the attention indicator and the feedback text word vector of the second solar cell usage feedback on the attention indicator. The common description array of the details of the details is used to reflect the common value of the details of the details between the feedback text word vector of the first solar cell usage feedback on the attention indicator and the feedback text word vector of the second solar cell usage feedback on the attention indicator.

[0009] The common description array of the details of the several indicators of interest is globally organized to obtain the global word vector of the feedback text.

[0010] In some optional embodiments, the step of globally organizing the common description array of the details of the several attention indicators to obtain the global word vector of the feedback text includes:

[0011] The common description arrays of details of several key indicators are combined to obtain the combined feedback text word vectors;

[0012] Vector mining is performed on the combined feedback text word vectors to obtain the global word vectors of the feedback text.

[0013] In some optional embodiments, the step of globally organizing the feedback text word vectors of the first solar cell usage feedback on several key indicators and the feedback text word vectors of the second solar cell usage feedback on several key indicators to obtain global feedback text word vectors reflecting the commonality values ​​of user needs in the first solar cell usage feedback and user needs details in the second solar cell usage feedback includes:

[0014] The first feedback text word vector is obtained by combining the feedback text word vectors of several attention indicators on the first solar cell.

[0015] The second solar cell is used to combine the feedback text word vectors of several attention indicators to obtain the second feedback text word vector.

[0016] By combining the first feedback text word vector and the second feedback text word vector, a feedback text word vector is determined that reflects the common value of the details of the matter between the first feedback text word vector and the second feedback text word vector, and this is used as the global word vector of the feedback text.

[0017] In some optional embodiments, before the step of globally organizing the feedback text word vectors of the first solar cell usage feedback on several attention indicators and the feedback text word vectors of the second solar cell usage feedback on several attention indicators to obtain the global word vectors of feedback text reflecting the commonality values ​​of user needs in the first solar cell usage feedback and user needs details in the second solar cell usage feedback, the method further includes: transferring the feedback text word vectors of the first solar cell usage feedback on several attention indicators and the feedback text word vectors of the second solar cell usage feedback on several attention indicators to the same text description relationship network and performing downsampling operation;

[0018] The step of globally organizing the feedback text word vectors of the first solar cell's use in several key indicators and the feedback text word vectors of the second solar cell's use in several key indicators to obtain the global word vectors of the feedback text includes: globally organizing the feedback text word vectors of the first solar cell's use in several key indicators after the downsampling operation and the feedback text word vectors of the second solar cell's use in several key indicators after the downsampling operation to obtain the global word vectors of the feedback text.

[0019] In some optional embodiments, the step of globally organizing the feedback text word vectors of the first solar cell usage feedback on several attention indicators and the feedback text word vectors of the second solar cell usage feedback on several attention indicators to obtain global word vectors of feedback texts reflecting the commonality values ​​of user needs in the first solar cell usage feedback and user needs details in the second solar cell usage feedback, and determining whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords by combining the global word vectors of feedback texts, is executed by a pre-tuned AI neural network.

[0020] In some optional embodiments, the AI ​​neural network is obtained by debugging using the following approach:

[0021] Obtain the first and second certified user feedback pairs. The commonality values ​​of the emotional word vectors of the two sets of solar cell user feedback in the first and second certified user feedback pairs are both higher than the specified commonality judgment value of the details. The user needs in the two sets of solar cell user feedback in the first certified user feedback pair are the same user needs, and the user needs in the two sets of solar cell user feedback in the second certified user feedback pair are different user needs.

[0022] The AI ​​neural network is obtained by debugging the original neural network specified by the first certified feedback tuple and the second certified feedback tuple.

[0023] In some alternative embodiments, the second solar cell is determined using feedback combined with the following approach:

[0024] Determine the common value of the details of the emotional word vectors of the demand items in the feedback of the use of the first solar cell and the emotional word vectors of the demand items in the feedback of the use of each group of solar cells marked with completion demand keywords;

[0025] The top X groups of solar cell usage feedback with the highest common value of the details mentioned above will be used as the second solar cell usage feedback, where X is a positive integer.

[0026] In some optional embodiments, determining whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords by combining the global word vectors of the feedback text includes:

[0027] For each group of second solar cell usage feedback, the probability score of matching the first solar cell usage feedback and each group of second solar cell usage feedback being the same demand keyword is determined by combining the global word vector of the feedback text.

[0028] Based on the fact that the highest value in the probability score is greater than the specified probability score, the first solar cell usage feedback and the second solar cell usage feedback corresponding to the highest value are determined to be matching the same demand keyword.

[0029] Based on the fact that the highest value in the probability score is less than the specified probability score, a derived demand keyword is generated, and the feedback on the use of the first solar cell is marked using the derived demand keyword.

[0030] Under some independent design approaches, the word vectors for the usage scenario experience are determined by combining the following ideas:

[0031] The detection stage in the user feedback detection process that is configured with the solar cell usage feedback crawling module is taken as a scene unit. The directed transmission feature between two scene units whose battery usage habit change value is less than a specified change value and / or whose cumulative battery usage period is less than a specified period is taken as a connection reference to generate a usage scene feature relationship network.

[0032] The target feature array of the scene unit is determined by combining the aforementioned usage scenario feature relationship network;

[0033] The target feature array is used as the usage scenario experience word vector of solar cell usage feedback crawling module, which is matched with the scene unit.

[0034] Under some independent design approaches, the target feature array of the scene unit is determined by combining the aforementioned usage scenario feature relationship network, including:

[0035] At least one set of usage scenario features is selected from the usage scenario feature relationship network using non-constraint rules, and each set of usage scenario feature features corresponds to a set of scenario unit queues.

[0036] The basic feature array of the scene units in the usage scenario feature relationship network is determined by combining the scene unit queues corresponding to the not less than one set of usage scenario feature sets;

[0037] For one of the scene units in the usage scenario feature relationship network, the basic feature array of the scene unit is iteratively optimized using the basic feature array of the associated scene units to obtain the target feature array of the scene unit.

[0038] Under some independent design approaches, determining the scene unit queue from the usage scenario feature relationship network through non-constraint rules includes:

[0039] During unconstrained filtering, for each current scene unit, the next scene unit corresponding to the current scene unit after one round of unconstrained filtering is determined by the following approach:

[0040] Determine the scene influence index of the current scene unit and each associated scene unit of the current scene unit. The scene influence index is used to reflect the reliability of the correlation coefficient between the current scene unit and the associated scene units. The scene influence index is determined by combining the battery usage habit change value of the current scene unit and each associated scene unit, and the cumulative battery usage time period of the current scene unit and each associated scene unit.

[0041] Based on the scenario impact index, a probability score list is determined pointing from the current scenario unit to each associated scenario unit;

[0042] The next scene unit is determined from the associated scene units by combining the probability score list.

[0043] In this embodiment of the invention, during the process of matching demand keywords in solar cell usage feedback, the word vectors of the feedback text (emotional word vectors of demand items, word vectors of usage scenario experience) of the first solar cell usage feedback to be marked with demand keywords and the second solar cell usage feedback to be marked with demand keywords can be obtained in several attention indicators. Then, the word vectors of the feedback text of the first solar cell usage feedback and the second solar cell usage feedback in several attention indicators are globally organized to obtain global word vectors of feedback text that reflect the commonality values ​​of the details of user demand items in the first solar cell usage feedback and the second solar cell usage feedback. Then, the global word vectors of feedback text are combined to determine whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords, so as to perform demand keyword matching / marking on the first solar cell usage feedback. By globally organizing the word vectors of feedback texts for different attention indicators, we can comprehensively analyze the contribution of these word vectors to the tagging of demand keywords, as well as the mutual influence between them. This allows the globally organized global word vectors of feedback texts to accurately and reliably reflect the commonalities of user needs in the two sets of solar cell usage feedback, thereby ensuring the accuracy and reliability of tagging demand keywords by combining the global word vectors of feedback texts. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0045] The methods, systems, and / or procedures shown in the accompanying drawings will be further described with reference to exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein reference numerals in the various views of the drawings represent similar mechanisms.

[0046] Figure 1 This is a block diagram illustrating an exemplary application scenario of a user demand analysis method for solar cells, as shown in some embodiments of the present invention.

[0047] Figure 2This is a schematic diagram of the hardware and software components of an exemplary user requirements analysis system according to some embodiments of the present invention.

[0048] Figure 3 This is a flowchart illustrating an exemplary user demand analysis method and / or process applied to solar cells, according to some embodiments of the present invention. Detailed Implementation

[0049] To better understand the above technical solutions, the technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solutions of the present invention, rather than limitations on the technical solutions of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0050] In the following detailed description, numerous specific details are illustrated by example to provide a comprehensive understanding of the relevant guidance. However, it will be apparent to those skilled in the art that the invention can be practiced without these details. In other instances, well-known methods, procedures, systems, components, and / or circuits have been described at a relatively high level without detail to avoid unnecessarily obscuring aspects of the invention.

[0051] These and other characteristics, the functions disclosed in the current application, the methods of execution, the functions of related elements in the structure, the combination of components, and the economics of production, will become more apparent in consideration of the following description with reference to the accompanying drawings, all of which form part of this invention. However, it should be clearly understood that the drawings are for illustrative and descriptive purposes only and are not intended to limit the scope of the invention. It should be understood that these drawings are not drawn to scale.

[0052] This invention uses flowcharts to illustrate the execution process performed by a system according to embodiments of the invention. It should be clearly understood that the execution processes in the flowchart may not be executed sequentially. Instead, these execution processes may be executed in reverse order or simultaneously. Additionally, at least one other execution process may be added to the flowchart. One or more execution processes may be deleted from the flowchart.

[0053] Figure 1 This is a block diagram of an exemplary application architecture 300 for a user requirements analysis method according to some embodiments of the present invention. The application architecture 300 for the user requirements analysis method may include a user requirements analysis system 100 and a solar cell user client 200.

[0054] In some embodiments, such as Figure 2 As shown, the user requirements analysis system 100 may include a processing engine 110, a network module 120, and a memory 130, with the processing engine 110 and the memory 130 communicating through the network module 120.

[0055] Processing engine 110 can process relevant information and / or data to perform one or more functions described in this invention. Network module 120 can facilitate the exchange of information and / or data. For example, network module 120 may include wired or wireless network access points, such as base stations and / or network access points. Memory 130 is used to store programs, which the processing engine 110 executes upon receiving execution instructions. It is understood that... Figure 2 The structure shown is for illustrative purposes only; the user requirements analysis system 100 may also include... Figure 2 The more or fewer components shown, or having the same Figure 2 The different configurations shown. Figure 2 The components shown can be implemented using hardware, software, or a combination thereof.

[0056] Figure 3 This is a flowchart illustrating an exemplary user demand analysis method and / or process applied to solar cells according to some embodiments of the present invention. The user demand analysis method applied to solar cells is... Figure 1 The user requirements analysis system 100 may further include the technical solutions described below.

[0057] Step 100: Sequentially obtain the first solar cell usage feedback to be marked with demand keywords and the second solar cell usage feedback to be marked with demand keywords, in the feedback text word vectors of several attention indicators.

[0058] In this embodiment of the invention, the feedback text word vectors of the plurality of attention indicators include at least: emotional word vectors of the demand and word vectors of the usage scenario experience.

[0059] Furthermore, the emotional word vectors for demand items are used to reflect the emotional field of user demand items in the user demand feedback of solar cell usage, and the emotional word vectors for usage scenario experience are used to reflect the scenario-based usage experience field of user demand items in the user demand feedback of solar cell usage.

[0060] For example, the first solar cell usage feedback to be tagged with demand keywords can be understood as usage feedback to be tagged / categorized, while the second solar cell usage feedback to be tagged with demand keywords can be understood as a reference for already tagged usage feedback. Solar cell usage feedback can be text feedback data, voice feedback data, etc., without limitation here. Attention indicators can be understood as attention dimensions, such as quality attention dimensions, safety attention dimensions, maintenance attention dimensions, etc. Feedback text word vectors can be the feedback features of solar cell usage feedback. The demand-related emotional word vectors and usage scenario experience word vectors correspond to the user's usage demand emotional features and the solar cell's usage scenario experience features, respectively. The usage demand emotional features involve the user's emotional polarity information, while the usage scenario experience features involve the solar cell's usage interaction.

[0061] Step 200: Globally organize the word vectors of the feedback text on several key indicators of the first solar cell usage feedback and the word vectors of the feedback text on several key indicators of the second solar cell usage feedback to obtain global word vectors of feedback text that reflect the commonality values ​​of the details of user needs in the first solar cell usage feedback and user needs in the second solar cell usage feedback.

[0062] In some possible examples, global organization can be understood as fusion / integration, common values ​​of user needs details can be understood as feature similarity corresponding to user needs, and global word vectors of feedback text can be understood as word vectors of the fused text.

[0063] In some alternative approaches, the global organization of the feedback text word vectors of the first solar cell usage feedback on several key indicators and the second solar cell usage feedback on several key indicators, as described in step 200, to obtain global feedback text word vectors that reflect the commonality values ​​of user needs in the first solar cell usage feedback and the user needs details in the second solar cell usage feedback, may include the technical solutions described in steps 210 and 220.

[0064] Step 210: For one of the several attention indicators, combine the feedback text word vectors of the first solar cell usage feedback on the attention indicator and the feedback text word vectors of the second solar cell usage feedback on the attention indicator to determine the common description array of the details of the attention indicator.

[0065] In some examples, the item detail commonality description array is used to reflect the item detail commonality values ​​between the first solar cell usage feedback on the feedback text word vector of the attention indicator and the second solar cell usage feedback on the feedback text word vector of the attention indicator.

[0066] Step 220: Globally organize the common description array of the details of the several indicators of interest to obtain the global word vector of the feedback text.

[0067] Applied to steps 210 and 220, it can accurately calculate the common description array of the details of the matter, thereby ensuring the integrity and reliability of the global word vector of the feedback text.

[0068] Under some possible design approaches, the step 220, which describes the global organization of the common description array of the details of the several indicators of interest to obtain the global word vector of the feedback text, may include the following: combining the common description array of the details of the several indicators of interest to obtain the combined word vector of the feedback text; and performing vector mining on the combined word vector of the feedback text to obtain the global word vector of the feedback text.

[0069] In other possible design approaches, the global organization of the feedback text word vectors for the first solar cell usage feedback across several key indicators and the second solar cell usage feedback across several key indicators, as described in step 200, to obtain a global feedback text word vector reflecting the commonality of user needs details in the first and second solar cell usage feedback, can be achieved as follows: Combine the feedback text word vectors for the first solar cell usage feedback across several key indicators to obtain a first feedback text word vector; combine the feedback text word vectors for the second solar cell usage feedback across several key indicators to obtain a second feedback text word vector; combine the first and second feedback text word vectors to determine a feedback text word vector reflecting the commonality of details between the first and second feedback text word vectors, and use this as the global feedback text word vector. This design considers both current and reference usage feedback, thereby ensuring the richness and completeness of the global feedback text word vector.

[0070] In some other possible embodiments, before the global organization of the feedback text word vectors of the first solar cell usage feedback on several attention indicators and the second solar cell usage feedback on several attention indicators as described in step 200 to obtain the global feedback text word vectors reflecting the commonality values ​​of user needs in the first solar cell usage feedback and user needs details in the second solar cell usage feedback, the following may also be included: transferring the feedback text word vectors of the first solar cell usage feedback on several attention indicators and the feedback text word vectors of the second solar cell usage feedback on several attention indicators to the same text description relationship network and performing downsampling operation.

[0071] It is understandable that the text description relationship network corresponds to the feature space or vector space, and the downsampling operation can be understood as dimensionality reduction or simplification. Based on this, the global word vectors of the feedback texts of the first solar cell usage feedback on several key indicators and the second solar cell usage feedback on several key indicators are globally processed to obtain the global word vectors of the feedback texts. This involves globally processing the feedback text word vectors of the first solar cell usage feedback on several key indicators after the downsampling operation and the second solar cell usage feedback on several key indicators after the downsampling operation to obtain the global word vectors of the feedback texts. This design ensures the simplification of the global word vectors of the feedback texts while maintaining feature recognition accuracy, thereby improving the basis for subsequent word vector analysis.

[0072] Step 300: Combine the global word vector of the feedback text to determine whether the feedback on the use of the first solar cell and the feedback on the use of the second solar cell match the same requirement keywords.

[0073] In this embodiment of the invention, solar cell usage feedback matching the same demand keywords refers to solar cell usage feedback addressing the same user demand. These demand keywords can include terms such as "usage guidelines," "energy conservation and emission reduction," and "visual tutorials."

[0074] In some cases, the way feedback on solar cell usage for the same user needs may be expressed may differ. Through the above-mentioned keyword matching analysis, we can analyze and process solar cell usage feedback in different ways and then perform clustering to improve the efficiency of user needs analysis and avoid excessive and ineffective analysis of solar cell usage feedback in different ways for the same user needs.

[0075] Under some independent design approaches, the above-mentioned steps of globally organizing the feedback text word vectors of the first solar cell usage feedback on several key indicators and the feedback text word vectors of the second solar cell usage feedback on several key indicators to obtain global word vectors of feedback text reflecting the commonality values ​​of user needs in the first solar cell usage feedback and the second solar cell usage feedback, and determining whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords by combining the global word vectors of feedback text, are executed by a pre-tuned AI neural network, which can be an RNN (Recurrent Neural Network) or a GCN (Graph Convolutional Neural Network).

[0076] Based on the above, the AI ​​neural network is obtained through debugging using the following approach: A first and a second certified usage feedback pair are obtained. The commonality values ​​of the emotional word vectors for the demand items in both sets of solar cell usage feedback in the first and second certified usage feedback pairs are higher than the specified commonality judgment value for item details. The user demand items in the two sets of solar cell usage feedback in the first certified usage feedback pair are the same, while the user demand items in the two sets of solar cell usage feedback in the second certified usage feedback pair are different. The AI ​​neural network is then obtained by debugging the original neural network specified by the first and second certified usage feedback pairs. For example, a certified usage feedback pair can be understood as a sample usage feedback pair.

[0077] In other possible design approaches, the second solar cell usage feedback is determined by combining the following approach: determining the commonality value of the emotional word vectors of the demand items in the first solar cell usage feedback and the emotional word vectors of the demand items in each group of solar cell usage feedback marked with completion demand keywords; selecting the top X groups of solar cell usage feedback with the highest commonality value of the demand details as the second solar cell usage feedback, where X is a positive integer. Based on this, step 300, by combining the global word vectors of the feedback text, determines whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords, which may include the technical solutions described in steps 310-330.

[0078] Step 310: For each group of second solar cell usage feedback, determine the probability score that the first solar cell usage feedback and each group of second solar cell usage feedback match the same demand keyword by combining the global word vector of the feedback text.

[0079] Step 320: Based on the fact that the highest value in the probability score is greater than the specified probability score, determine that the first solar cell usage feedback and the second solar cell usage feedback corresponding to the highest value are matching keywords with the same demand.

[0080] Step 330: Based on the fact that the highest value in the probability score is less than the specified probability score, generate a derived demand keyword, and use the derived demand keyword to mark the feedback on the use of the first solar cell.

[0081] It is understandable that the probability score corresponds to the probability value that the first solar cell usage feedback and each group of second solar cell usage feedback match the same demand keyword. Based on this, flexible labeling processing can be performed through the probability score.

[0082] Under some independent design approaches, the usage scenario experience word vectors are determined by combining the following ideas: The detection stage in the user feedback detection process that is configured with the solar cell usage feedback crawling module is taken as a scenario unit; the directed transmission features (such as association information) between two scenario units whose battery usage habit change value (e.g., user operation characteristic change) is less than a specified change value and / or whose cumulative battery usage time is less than a specified time period are used as connection references to generate a usage scenario feature relationship network; the target feature array of the scenario unit is determined by combining the usage scenario feature relationship network; the target feature array is used as the usage scenario experience word vectors of the solar cell usage feedback crawling module (e.g., data crawler) that are matched with the scenario unit.

[0083] Under some independent design approaches, determining the target feature array of a scenario unit by combining the usage scenario feature relationship network can include the following: Selecting at least one set of usage scenario features from the usage scenario feature relationship network using non-constrained rules (such as a random strategy), with each set of usage scenario features corresponding to a queue of scenario units; determining the basic feature array of scenario units in the usage scenario feature relationship network by combining the queues of scenario units corresponding to the at least one set of usage scenario features; for one scenario unit in the usage scenario feature relationship network, iteratively optimizing the basic feature array of the scenario unit using the basic feature arrays of its associated scenario units to obtain the target feature array of the scenario unit. This design ensures the accuracy and timeliness of the target feature array.

[0084] Under some independent design approaches, determining the scene unit queue from the usage scenario feature relationship network through non-constrained rules includes: during non-constrained filtering, for each current scene unit, determining the next scene unit corresponding to the current scene unit after one round of non-constrained filtering based on the following approach: determining the scene influence index of the current scene unit and each associated scene unit of the current scene unit, the scene influence index being used to reflect the reliability of the correlation coefficient between the current scene unit and the associated scene units; wherein, the scene influence index is determined by combining the battery usage habit change value of the current scene unit and each associated scene unit, and the cumulative battery usage time period of the current scene unit and each associated scene unit; determining a probability score list pointing from the current scene unit to each associated scene unit based on the scene influence index; and determining the next scene unit from each associated scene unit based on the probability score list. Here, a scene unit can be understood as a modular local scene. Based on the above, this design ensures the integrity of the scene unit queue.

[0085] Understandably, by implementing steps 100-300, during the process of matching demand keywords in the feedback on solar cell usage, the word vectors of the first solar cell usage feedback to be marked with demand keywords and the second solar cell usage feedback to be marked with demand keywords can be obtained in terms of several attention indicators (emotional word vectors of demand items, word vectors of usage scenario experience). Then, the word vectors of the first and second solar cell usage feedback in terms of several attention indicators are globally organized to obtain global word vectors of feedback text that reflect the commonality values ​​of the details of user demand items in the first and second solar cell usage feedback. Then, the global word vectors of feedback text are combined to determine whether the first and second solar cell usage feedback match the same demand keywords, so as to perform demand keyword matching / marking on the first solar cell usage feedback. By globally organizing the word vectors of feedback texts for different attention indicators, we can comprehensively analyze the contribution of these word vectors to the tagging of demand keywords, as well as the mutual influence between them. This allows the globally organized global word vectors of feedback texts to accurately and reliably reflect the commonalities of user needs in the two sets of solar cell usage feedback, thereby ensuring the accuracy and reliability of tagging demand keywords by combining the global word vectors of feedback texts.

[0086] The above-described content disclosed in the embodiments of this invention is clear and complete to those skilled in the art. It should be understood that the process by which those skilled in the art derive and analyze the unexplained technical terms based on the above-disclosed content is based on the content described in this invention, and therefore the above content is not a judgment of the inventiveness of the overall solution.

[0087] The basic concepts have been described above. It is obvious that the detailed disclosure above is merely illustrative and does not constitute a limitation of the present invention. Although not explicitly stated herein, various modifications, improvements, and alterations can be made to the present invention by those skilled in the art. Such modifications, improvements, and alterations are suggested in this invention and therefore remain within the spirit and scope of the exemplary embodiments of the present invention.

[0088] Furthermore, this invention uses specific terminology to describe embodiments of the invention. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of the invention. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different parts of this specification do not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in at least one embodiment of the invention can be appropriately combined.

[0089] Furthermore, it will be understood by those skilled in the art that various aspects of the present invention can be described and illustrated through several patentable types or situations, including any new and useful combination of processes, machines, products, or substances, or any new and useful improvements thereof. Accordingly, various aspects of the present invention can be implemented entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. All of the above hardware or software can be referred to as a “unit,” “component,” or “system.” Moreover, various aspects of the present invention can be embodied as a computer product located on at least one computer-readable medium, said product comprising computer-readable program code.

[0090] A computer-readable signal medium may contain a propagated data signal containing computer program encoding, for example, on baseband or as part of a carrier wave. This propagated signal may take various forms, including electromagnetic, optical, and so on, or suitable combinations thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, apparatus, or device to enable communication, propagation, or transmission of a program for use. The program encoding located on the computer-readable signal medium can be propagated through any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0091] The computer program code required for the execution of various aspects of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., or similar conventional programming languages ​​such as the "C" programming language, Visual Basic, Fortran2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby, and Groovy, or other programming languages. The program code can be executed entirely on the user's computer, or as a standalone software package on the user's computer, or partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer via any network, such as a local area network (LAN) or wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as Software as a Service (SaaS).

[0092] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numerals, or other names described in this invention are not intended to limit the order of the processes and methods of this invention. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the additional claims are not limited to the disclosed embodiments. Rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments of this invention. For example, while the system components described above can be implemented by hardware devices, they can also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.

[0093] It should also be understood that, in order to simplify the description of the invention and thus aid in the understanding of at least one embodiment, multiple features may sometimes be grouped into a single embodiment, drawing, or description thereof in the foregoing description of the embodiments of the invention. However, this method of disclosure does not imply that the subject matter of the invention requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiment disclosed above.

Claims

1. A user demand analysis method applied to solar cells, characterized in that, The method, applied to a user requirements analysis system, includes: The system sequentially obtains the first solar cell usage feedback to be marked with demand keywords and the second solar cell usage feedback after the demand keywords have been marked, along with the feedback text word vectors for several key indicators. The feedback text word vectors for these key indicators include at least: demand item sentiment word vectors and usage scenario experience word vectors. The demand item sentiment word vectors reflect the user demand sentiment field of the user demand item in the solar cell usage feedback, and the usage scenario experience word vectors reflect the scenario-based usage experience field of the user demand item in the solar cell usage feedback. The word vectors of the feedback texts on several key indicators of the first solar cell usage feedback and the word vectors of the feedback texts on several key indicators of the second solar cell usage feedback are globally organized to obtain global word vectors of feedback texts that reflect the commonality of details of user needs in the first solar cell usage feedback and user needs in the second solar cell usage feedback. The global word vector of the feedback text is used to determine whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords; wherein, solar cell usage feedback that matches the same demand keywords is solar cell usage feedback for the same user demand.

2. The method according to claim 1, characterized in that, The process of globally organizing the word vectors of the feedback text on several key indicators of the first solar cell usage feedback and the word vectors of the feedback text on several key indicators of the second solar cell usage feedback, to obtain global word vectors of feedback text reflecting the commonality of details of user needs in the first solar cell usage feedback and the second solar cell usage feedback, includes: For one of the several attention indicators, a common description array of the details of the attention indicator is determined by combining the feedback text word vector of the first solar cell usage feedback on the attention indicator and the feedback text word vector of the second solar cell usage feedback on the attention indicator. The common description array of the details of the details is used to reflect the common value of the details of the details between the feedback text word vector of the first solar cell usage feedback on the attention indicator and the feedback text word vector of the second solar cell usage feedback on the attention indicator. The common description array of the details of the several indicators of interest is globally organized to obtain the global word vector of the feedback text.

3. The method according to claim 2, characterized in that, The process of globally organizing the common description array of the details of the several indicators of interest to obtain the global word vector of the feedback text includes: The common description arrays of details of several key indicators are combined to obtain the combined feedback text word vectors; Vector mining is performed on the combined feedback text word vectors to obtain the global word vectors of the feedback text.

4. The method according to claim 1, characterized in that, The process of globally organizing the word vectors of the feedback text on several key indicators of the first solar cell usage feedback and the word vectors of the feedback text on several key indicators of the second solar cell usage feedback, to obtain global word vectors of feedback text reflecting the commonality of details of user needs in the first solar cell usage feedback and the second solar cell usage feedback, includes: The first feedback text word vector is obtained by combining the feedback text word vectors of several attention indicators on the first solar cell. The second solar cell is used to combine the feedback text word vectors of several attention indicators to obtain the second feedback text word vector. By combining the first feedback text word vector and the second feedback text word vector, a feedback text word vector is determined that reflects the common value of the details of the matter between the first feedback text word vector and the second feedback text word vector, and this is used as the global word vector of the feedback text.

5. The method according to claim 1, characterized in that, Before the step of globally organizing the feedback text word vectors of the first solar cell usage feedback on several key indicators and the feedback text word vectors of the second solar cell usage feedback on several key indicators to obtain the global word vectors of feedback text reflecting the commonality values ​​of user needs in the first solar cell usage feedback and user needs in the second solar cell usage feedback, the method further includes: transferring the feedback text word vectors of the first solar cell usage feedback on several key indicators and the feedback text word vectors of the second solar cell usage feedback on several key indicators to the same text description relationship network and performing downsampling operation; The step of globally organizing the feedback text word vectors of the first solar cell's use in several key indicators and the feedback text word vectors of the second solar cell's use in several key indicators to obtain the global word vectors of the feedback text includes: globally organizing the feedback text word vectors of the first solar cell's use in several key indicators after the downsampling operation and the feedback text word vectors of the second solar cell's use in several key indicators after the downsampling operation to obtain the global word vectors of the feedback text.

6. The method according to claim 1, characterized in that, The step of globally organizing the word vectors of the feedback text on several key indicators of the first solar cell usage feedback and the word vectors of the feedback text on several key indicators of the second solar cell usage feedback to obtain global word vectors of feedback text that reflect the commonality of user needs in the first solar cell usage feedback and the second solar cell usage feedback, and determining whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords by combining the global word vectors of feedback text, is executed by a pre-tuned AI neural network.

7. The method according to claim 6, characterized in that, The AI ​​neural network was obtained through debugging using the following approach: Obtain the first and second certified user feedback pairs. The commonality values ​​of the emotional word vectors of the two sets of solar cell user feedback in the first and second certified user feedback pairs are both higher than the specified commonality judgment value of the details. The user needs in the two sets of solar cell user feedback in the first certified user feedback pair are the same user needs, and the user needs in the two sets of solar cell user feedback in the second certified user feedback pair are different user needs. The AI ​​neural network is obtained by debugging the original neural network specified by the first certified feedback tuple and the second certified feedback tuple.

8. The method according to claim 1, characterized in that, The second solar cell was determined using feedback and the following approach: Determine the common value of the details of the emotional word vectors of the demand items in the feedback of the use of the first solar cell and the emotional word vectors of the demand items in the feedback of the use of each group of solar cells marked with completion demand keywords; The top X groups of solar cell usage feedback with the highest common value of the details mentioned above will be used as the second solar cell usage feedback, where X is a positive integer.

9. The method according to claim 8, characterized in that, The step of determining whether the first solar cell usage feedback and the second solar cell usage feedback match the same demand keywords by combining the global word vectors of the feedback text includes: For each group of second solar cell usage feedback, the probability score of matching the first solar cell usage feedback and each group of second solar cell usage feedback being the same demand keyword is determined by combining the global word vector of the feedback text. Based on the fact that the highest value in the probability score is greater than the specified probability score, the first solar cell usage feedback and the second solar cell usage feedback corresponding to the highest value are determined to be matching the same demand keyword. Based on the fact that the highest value in the probability score is less than the specified probability score, a derived demand keyword is generated, and the feedback on the use of the first solar cell is marked using the derived demand keyword.

10. A user requirements analysis system, characterized in that, The method includes a processing engine, a network module, and a memory, wherein the processing engine and the memory communicate through the network module, and the processing engine is used to read and run a computer program from the memory to implement the method of any one of claims 1-9.

Images