Method and device for determining customer portrait, electronic equipment and storage medium

Abstract

This disclosure provides a method, apparatus, electronic device, and storage medium for determining customer profiles. The method involves acquiring pre-training corpora from a pre-training corpus, the pre-training corpus containing at least two pre-training corpora; determining the target size of the at least two pre-training corpora; scaling the different pre-training corpora in the pre-training corpus according to the target size; merging the target corpus with the scaled-up at least two pre-training corpora to obtain at least two target corpora; and inputting the at least two target corpora into a unary recognition model to obtain the customer profile recognition result of the target corpus output by the unary recognition model. The unary recognition model is used to identify individual words in the corpus. Compared with related technologies, this disclosure, by merging and processing the target corpus, identifies the customer profile of the target corpus, and the unary recognition model can effectively improve the accuracy of customer profile recognition.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a method and apparatus for determining customer profiles, electronic devices, and storage media. Background Technology

[0002] The widespread use of the internet has led to an exponential increase in the amount of textual information we encounter, much of which has unknown origins. Personal language style refers to a person's habitual expression within a given language. To identify the origin of text from unknown customers, we can determine its attribution by comparing the language style characteristics of the unknown customer with those of known customers. However, the accuracy of current customer profiling methods still needs improvement. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, and storage medium for determining customer profiles. Its main purpose is to solve the problem of low accuracy in customer profile recognition methods.

[0004] According to a first aspect of this disclosure, a method for determining a customer profile is provided, comprising:

[0005] Obtain pre-training data from a pre-training corpus, wherein the pre-training corpus contains at least two pre-training data sets;

[0006] Determine the target size of the at least two pre-training corpora, and scale the different pre-training corpora in the pre-training corpus according to the target size;

[0007] The corpus to be determined is merged with at least two pre-trained corpora after scaling to obtain at least two target corpora;

[0008] The at least two target corpora are input into a unary recognition model to obtain the customer profile recognition result of the corpus to be determined output by the unary recognition model, wherein the unary recognition model is used to identify individual words in the corpus.

[0009] Optionally, scaling the different pre-training corpora in the pre-training corpus according to the target size includes:

[0010] Based on the target size, determine the scaling factor of each of the at least two pre-training corpora;

[0011] Scale according to the scaling coefficients of each of the at least two pre-trained corpora.

[0012] Optionally, merging the corpus to be determined with at least two scaled pre-trained corpora to obtain at least two target corpora includes:

[0013] A preset overlay algorithm is invoked to overlay the corpus to be determined with at least two expanded pre-training corpora.

[0014] The result of the superposition process is subjected to sentence smoothing to obtain at least two target corpora, in which the probability of each word appearing is not 0.

[0015] Optionally, obtaining the customer profile recognition result of the unary recognition model outputting the corpus to be determined includes:

[0016] Within the unary recognition model, determine the word frequency of each word in the corpus to be determined in each target corpus;

[0017] The probability value of each sentence in the corpus to be determined is calculated based on the word frequency in each target corpus;

[0018] The customer profile recognition result is determined based on the probability value of each sentence.

[0019] Optionally, the determination of the word frequency of each word in the corpus to be determined in each target corpus within the unary recognition model is achieved by the following formula:

[0020] p(w i ) = count(w i ) / count(words)

[0021] count(w i ) represents the word w i The count(words) represents the total number of times all words appear in the target corpus.

[0022] Optionally, the step of calculating the probability value of each sentence in the corpus to be determined based on the word frequency in each target corpus is achieved by the following formula:

[0023] p(s)=p(w1)p(w2)p(w3)...p(w n )

[0024] or,

[0025] log p(s)=log p(w1)+logp(w2)+log p(w3)+...+log p(w n )

[0026] w i 's' represents the words or phrases that appear in the target corpus, and 's' represents the sentences in the corpus to be determined.

[0027] According to a second aspect of this disclosure, a customer profiling determination apparatus is provided, comprising:

[0028] An acquisition unit is used to acquire pre-training corpora from a pre-training corpus, wherein the pre-training corpus contains at least two pre-training corpora.

[0029] The scaling unit is used to determine the target size of the at least two pre-training corpora and to scale the different pre-training corpora of the pre-training corpus according to the target size.

[0030] The merging unit is used to merge the corpus to be determined with at least two scaled pre-training corpora to obtain at least two target corpora.

[0031] The determining unit is used to input the at least two target corpora into a unary recognition model and obtain the customer profile recognition result of the corpora to be determined output by the unary recognition model, wherein the unary recognition model is used to identify individual words in the corpora.

[0032] Optionally, the telescopic unit includes:

[0033] The first determining module is used to determine the scaling factor of each of the at least two pre-training corpora based on the target size;

[0034] A scaling module is used to scale the data according to the scaling coefficients of the at least two pre-trained corpora.

[0035] Optionally, the merging unit includes:

[0036] The calling module is used to call a preset superposition algorithm to superimpose the corpus to be determined with at least two stretched pre-training corpora respectively;

[0037] The processing module is used to perform sentence smoothing on the result of the superposition processing to obtain the at least two target corpora, wherein the probability of each word appearing in the at least two target corpora is not 0.

[0038] Optionally, the determining unit includes:

[0039] The second determining module is used to determine the word frequency of each word in the corpus to be determined in each target corpus within the unary recognition model;

[0040] The calculation module is used to calculate the probability value of each sentence in the corpus to be determined in each target corpus based on the word frequency;

[0041] The third determining module is used to determine the customer profile recognition result based on the probability value of each sentence.

[0042] Optionally, the second determining module is implemented using the following formula:

[0043] p(w i ) = count(w i ) / count(words)

[0044] count(w i ) represents the word w i The count(words) represents the total number of times all words appear in the target corpus.

[0045] Optionally, the calculation module is implemented using the following formula:

[0046] p(s)=p(w1)p(w2)p(w3)...p(w n )

[0047] or,

[0048] log p(s)=log p(w1)+log p(w2)+log p(w3)+...+log p(w n )

[0049] w i 's' represents the words or phrases that appear in the target corpus, and 's' represents the sentences in the corpus to be determined.

[0050] According to a third aspect of this disclosure, an electronic device is provided, comprising:

[0051] At least one processor; and

[0052] A memory communicatively connected to the at least one processor; wherein,

[0053] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect above.

[0054] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are configured to cause the computer to perform the method described in the first aspect above.

[0055] According to a fifth aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0056] This disclosure provides a method, apparatus, electronic device, and storage medium for determining customer profiles. The method involves acquiring pre-training corpora from a pre-training corpus, wherein the pre-training corpus contains at least two pre-training corpora; determining the target size of the at least two pre-training corpora; scaling the different pre-training corpora in the pre-training corpus according to the target size; merging the target corpus with the scaled-up at least two pre-training corpora to obtain at least two target corpora; and inputting the at least two target corpora into a unary recognition model to obtain the customer profile recognition result of the target corpus output by the unary recognition model. The unary recognition model is used to identify individual words in the corpus. Compared with related technologies, this disclosure, by merging and processing the target corpus, identifies the customer profile of the target corpus, and the unary recognition model can effectively improve the accuracy of customer profile recognition.

[0057] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0058] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0059] Figure 1 A flowchart illustrating a method for determining a customer profile provided in an embodiment of this disclosure;

[0060] Figure 2 A schematic diagram of a text scaling process provided in an embodiment of this disclosure;

[0061] Figure 3 A flowchart illustrating the unary recognition model recognition method provided in this embodiment of the disclosure;

[0062] Figure 4 A schematic diagram of a customer profile determination device provided in an embodiment of this disclosure;

[0063] Figure 5 A schematic diagram of another customer profile determination device provided in an embodiment of this disclosure;

[0064] Figure 6 A schematic block diagram of an example electronic device 500 provided for embodiments of this disclosure. Detailed Implementation

[0065] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0066] The following description, with reference to the accompanying drawings, outlines a method, apparatus, electronic device, and storage medium for determining customer profiles according to embodiments of the present disclosure.

[0067] Figure 1 This is a flowchart illustrating a method for determining a customer profile provided in an embodiment of this disclosure.

[0068] like Figure 1 As shown, the method includes the following steps:

[0069] Step 101: Obtain pre-training corpus from the pre-training corpus, wherein the pre-training corpus contains at least two pre-training corpora.

[0070] In some embodiments, the following implementation may be adopted, but is not limited to: the pre-training corpus is text pre-processed by a Natural Language Toolkit (NLTK). NLTK is used to segment all the integrated text, count the number of words, and determine the size of each text; each processed text yields a corresponding pre-training corpus, and each pre-training corpus has a unique corresponding customer profile identifier. In this embodiment, NLTK segments the text into unary models and stores the processed pre-training corpus in a pre-training corpus, which includes a large number of pre-processed pre-training corpora. The text processed by NLTK forms at least two pre-training corpora. It should be noted that this disclosure uses NLTK as an example for illustration and does not constitute a limitation on which natural language processing tool is used in this disclosure.

[0071] Step 102: Determine the target size of the at least two pre-training corpora, and scale the different pre-training corpora in the pre-training corpus according to the target size.

[0072] Inconsistent corpus sizes can lead to discrepancies between the pre-training and target corpora, potentially causing significant frequency discrepancies between words in the target corpus and their corresponding word frequencies. To address this corpus imbalance, text scaling of the pre-training corpus is necessary to ensure that each pre-training corpus is of equal size, meaning the total number of words in the vocabulary is equal.

[0073] Step 103: Merge the corpus to be determined with at least two pre-trained corpora after scaling to obtain at least two target corpora.

[0074] Incomplete information leads to data sparsity. Due to the differences between the corpus to be determined and the pre-training corpus, and the finiteness of the pre-training corpus, it cannot be guaranteed that all words in the corpus to be determined have appeared in the pre-training corpus. This can result in a word having a probability of zero, regardless of the frequency of other preceding words, leading to a final probability product of zero. Therefore, a non-zero probability value must be assigned to all possible words to avoid this situation. To address the zero-probability problem caused by data sparsity, after text scaling of at least two pre-training corpora, the corpus to be determined is merged with each of the scaled corpora to obtain at least two target corpora; these are then processed using a pre-defined algorithm.

[0075] Step 104: Input the at least two target corpora into the unary recognition model and obtain the customer profile recognition result of the corpora to be determined output by the unary recognition model, wherein the unary recognition model is used to identify individual words in the corpora.

[0076] Based on the unary recognition model of this disclosure embodiment, the processed target corpus is input into the model. The unary recognition model calculates the word frequency of words in the corpus to be determined, and further obtains the probability value corresponding to the corpus to be determined, thereby judging the customer profile recognition result.

[0077] This disclosure provides a method for determining customer profiles. The method involves acquiring pre-training corpora from a pre-training corpus, wherein the pre-training corpus contains at least two pre-training corpora; determining the target size of the at least two pre-training corpora; scaling the different pre-training corpora in the pre-training corpus according to the target size; merging the target corpus with the scaled-up at least two pre-training corpora to obtain at least two target corpora; inputting the at least two target corpora into a unary recognition model to obtain the customer profile recognition result of the target corpus output by the unary recognition model, wherein the unary recognition model is used to identify individual words in the corpus. Compared with related technologies, this disclosure, by merging and processing the target corpus under the unary model, identifies the customer profile of the target corpus. This disclosure can improve the success rate of customer profile recognition.

[0078] For a more intuitive understanding of the text scaling process, please refer to [link / reference]. Figure 2 , Figure 2 This is a schematic diagram illustrating a text scaling process provided in an embodiment of this disclosure. The step of scaling different pre-training corpora in the pre-training corpus according to the target size includes:

[0079] Step 201: Determine the scaling factor of each of the at least two pre-training corpora according to the target size.

[0080] Based on a defined target size, which is the largest size among all pre-training corpora, a scaling factor is determined for each pre-training corpus, which is the ratio of the target size to the size of the pre-training corpus. The frequency of each word in the pre-training corpus is obtained by multiplying the word count by the scaling factor.

[0081] Step 202: Scale the data according to the scaling coefficients of the at least two pre-trained corpora.

[0082] Based on the scaling factor of each pre-training corpus, each pre-training corpus is scaled to obtain a scaled pre-training corpus. The word count of each word in the scaled pre-training corpus is the product of the frequency of each word in the original pre-training corpus and the scaling factor.

[0083] Furthermore, in one possible implementation of this embodiment, merging the corpus to be determined with at least two scaled pre-training corpora to obtain at least two target corpora includes:

[0084] A preset overlay algorithm is invoked to overlay the corpus to be determined with at least two expanded pre-training corpora.

[0085] The result of the superposition process is subjected to sentence smoothing to obtain at least two target corpora, in which the probability of each word appearing is not 0.

[0086] In the process of identifying customer profiles, each word in a sentence needs a non-zero probability value. If any word has a probability of zero, the entire calculation formula will result in zero. Therefore, a non-zero probability value must be assigned to all possible words to avoid this situation. In this embodiment, a preset overlay algorithm is used to add the vocabulary of the corpus to be determined to each expanded pre-training corpus vocabulary, resulting in a consistent size for each target corpus. The customer profile identification information of the target corpus is consistent with that of the pre-training corpus. Each target corpus is then processed using additive smoothing. It should be noted that the sentence smoothing process for the overlay processing result in this embodiment uses additive smoothing, which does not constitute a limitation on the algorithm used for smoothing.

[0087] To better illustrate how the unary recognition model identifies customer profiles, please refer to [link / reference]. Figure 3 , Figure 3This is a flowchart illustrating the unary recognition model recognition method provided in this embodiment. The step of obtaining the customer profile recognition result of the unary recognition model outputting the corpus to be determined includes:

[0088] Step 301: Determine the word frequency of each word in the corpus to be determined in each target corpus within the unary recognition model.

[0089] A unigram recognition model is used to determine the word frequency of each word in the target corpus. The probability of each word's occurrence is approximated by statistical word frequency.

[0090] Step 302: Calculate the probability value of each sentence in the corpus to be determined in each target corpus based on the word frequency.

[0091] The probability of each word's occurrence is approximated by its frequency, assuming that a word's occurrence is only related to itself and not to preceding words. In each target corpus, the probability value of each sentence in the corpus to be determined can be calculated from the probability values ​​of each word's occurrence; that is, the product of the probability values ​​of each word in a sentence is the sentence's probability value. A univariate model is used to calculate the probability value of each sentence in the corpus to be determined within each target corpus.

[0092] Step 303: Determine the customer profile recognition result based on the probability value of each sentence.

[0093] The probability value of each sentence in the undetermined corpus is calculated using a univariate model. This probability value is then compared with the probability values ​​of each sentence in all the undetermined corpora. The customer profile identification information of the target corpus corresponding to the highest probability value is obtained, and it is assumed that the undetermined corpus comes from the customer corresponding to the target corpus.

[0094] Furthermore, in one possible implementation of this embodiment, the determination of the word frequency of each word in the corpus to be determined within the unary recognition model is achieved by the following formula:

[0095] p(w i ) = count(w i ) / count(words)

[0096] count(w i ) represents the word w i The count(words) represents the total number of times all words appear in the target corpus.

[0097] In this embodiment of the disclosure, a unary recognition model is used to identify the attribution of customer profiles. The frequency of each word is used to approximate the probability of each word's occurrence.

[0098] Furthermore, in one possible implementation of this embodiment, the step of calculating the probability value of each sentence in the corpus to be determined based on the word frequency is achieved by the following formula:

[0099] p(s)=p(w1)p(w2)p(w3)...p(w n )

[0100] or,

[0101] log p(s)=log p(w1)+log p(w2)+log p(w3)+...+log p(w n )

[0102] w i 's' represents the words or phrases that appear in the target corpus, and 's' represents the sentences in the corpus to be determined.

[0103] If the occurrence of a word is only related to itself and not to the words that have already appeared, then the sentence s = w1w2w3……w n The probability of occurrence is the product of the probability values ​​of each word. In practice, if the test statement is too long, the result of the probability product will be very small, potentially even causing data overflow. Therefore, a logarithmic operation can be performed on the probability product.

[0104] Corresponding to the above-described method for determining customer profiles, this invention also proposes a device for determining customer profiles. Since the device embodiments of this invention correspond to the method embodiments described above, details not disclosed in the device embodiments can be referred to in the method embodiments, and will not be repeated here.

[0105] Figure 4 This is a schematic diagram of a customer profile determination device provided in an embodiment of the present disclosure, as shown below. Figure 4 As shown, it includes: an acquisition unit 41, a scaling unit 42, a merging unit 43, and a determination unit 44.

[0106] Acquisition unit 41 is used to acquire pre-training corpus in a pre-training corpus, wherein the pre-training corpus contains at least two pre-training corpora.

[0107] The scaling unit 42 is used to determine the target size of the at least two pre-training corpora and to scale the different pre-training corpora of the pre-training corpus according to the target size.

[0108] The merging unit 43 is used to merge the corpus to be determined with at least two expanded pre-training corpora to obtain at least two target corpora;

[0109] The determining unit 44 is used to input the at least two target corpora into a unary recognition model and obtain the customer profile recognition result of the corpora to be determined output by the unary recognition model, wherein the unary recognition model is used to identify individual words in the corpora.

[0110] Furthermore, in one possible implementation of this embodiment, such as Figure 5 As shown, the telescopic unit 42 includes:

[0111] The first determining module 421 is used to determine the scaling factor of each of the at least two pre-training corpora based on the target size;

[0112] The scaling module 422 is used to scale according to the scaling coefficients of the at least two pre-trained corpora.

[0113] Furthermore, in one possible implementation of this embodiment, such as Figure 5 As shown, the merging unit 43 includes:

[0114] Module 431 is used to call a preset superposition algorithm to superimpose the corpus to be determined with at least two stretched pre-trained corpora respectively.

[0115] The processing module 432 is used to perform sentence smoothing on the result of the superposition processing to obtain the at least two target corpora, wherein the probability of each word appearing in the at least two target corpora is not 0.

[0116] Furthermore, in one possible implementation of this embodiment, such as Figure 5 As shown, the determining unit 44 includes:

[0117] The second determining module 441 is used to determine the word frequency of each word in the corpus to be determined in each target corpus within the unary recognition model;

[0118] The calculation module 442 is used to calculate the probability value of each sentence in the corpus to be determined in each target prediction based on the word frequency;

[0119] The third determining module 443 is used to determine the customer profile recognition result based on the probability value of each sentence.

[0120] Furthermore, in one possible implementation of this embodiment, the second determining module 441 is implemented by the following formula:

[0121] p(w i ) = count(w i ) / count(words)

[0122] count(w i) represents the word w i The count(words) represents the total number of times all words appear in the target corpus.

[0123] Furthermore, in one possible implementation of this embodiment, the calculation module 442 is implemented using the following formula:

[0124] p(s)=p(w1)p(w2)p(w3)...p(w n )

[0125] or,

[0126] log p(s)=log p(w1)+log p(w2)+log p(w3)+...+log p(w n )

[0127] w i 's' represents the words or phrases that appear in the target corpus, and 's' represents the sentences in the corpus to be determined.

[0128] It should be noted that the foregoing explanation of the method embodiments also applies to the apparatus of this embodiment, and the principle is the same, so it is not limited in this embodiment.

[0129] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0130] Figure 6 A schematic block diagram of an example electronic device 500 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0131] like Figure 6As shown, device 500 includes a computing unit 501, which can perform various appropriate actions and processes based on a computer program stored in ROM (Read-Only Memory) 502 or a computer program loaded from storage unit 508 into RAM (Random Access Memory) 503. RAM 503 can also store various programs and data required for the operation of device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via bus 504. I / O (Input / Output) interface 505 is also connected to bus 504.

[0132] Multiple components in device 500 are connected to I / O interface 505, including: input unit 506, such as keyboard, mouse, etc.; output unit 507, such as various types of monitors, speakers, etc.; storage unit 508, such as disk, optical disk, etc.; and communication unit 509, such as network card, modem, wireless transceiver, etc. Communication unit 509 allows device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0133] The computing unit 501 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, CPUs (Central Processing Units), GPUs (Graphics Processing Units), various special-purpose AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, DSPs (Digital Signal Processors), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above, such as the customer profile determination method. For example, in some embodiments, the customer profile determination method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the methods described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the aforementioned customer profile determination method by any other suitable means (e.g., by means of firmware).

[0134] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, FPGAs (Field Programmable Gate Arrays), ASICs (Application-Specific Integrated Circuits), ASSPs (Application-Specific Standard Products), SOCs (System-on-Chips), CPLDs (Complex Programmable Logic Devices), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0135] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0136] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, RAM, ROM, EPROM (Electrically Programmable Read-Only Memory) or flash memory, optical fiber, CD-ROM (Compact Disc Read-Only Memory), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0137] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0138] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include LANs (Local Area Networks), WANs (Wide Area Networks), the Internet, and blockchain networks.

[0139] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem, addressing the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0140] It's important to note that artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It encompasses both hardware and software technologies. AI hardware technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, and big data processing. AI software technologies primarily include computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graph technologies.

[0141] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0142] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for determining a customer profile, characterized in that, include: Obtain pre-training data from a pre-training corpus, wherein the pre-training corpus contains at least two pre-training data sets; Determine the target size of the at least two pre-training corpora, and scale the different pre-training corpora in the pre-training corpus according to the target size; Obtain the corpus to be determined; The corpus to be determined is merged with at least two pre-trained corpora after scaling to obtain at least two target corpora; The at least two target corpora are input into a unary recognition model, and the word frequency of each word in the corpus to be determined in each target corpus is determined within the unary recognition model. The probability value of each sentence in the corpus to be determined is calculated based on the word frequency in each target corpus; Based on the probability value of each sentence, the target corpus corresponding to the maximum probability value is determined, and the customer profile identification information of the target corpus is obtained as the customer profile recognition result of the corpus to be determined. The unary recognition model is used to identify individual words in the corpus. The customer profile recognition results obtained from the unary recognition model outputting the corpus to be determined include: Wherein, the word frequency p( This can be achieved through the following formula: p( ) =count( ) / count(words) The probability value of each sentence is achieved using the following formula: p(s) = p( )p( )p( )...p( ) or, log p(s) = log p( )+log p( ) + log p( ) + ... + log p( ) in, This indicates the words or phrases appearing in the target corpus, where 's' represents sentences in the corpus to be determined; count( ) represents a word The count(words) represents the total number of times all words appear in the target corpus.

2. The method according to claim 1, characterized in that, The step of scaling the different pre-training corpora in the pre-training corpus according to the target size includes: Based on the target size, determine the scaling factor of each of the at least two pre-training corpora; Scale according to the scaling coefficients of each of the at least two pre-trained corpora.

3. The method according to claim 1, characterized in that, The step of merging the corpus to be determined with at least two scaled pre-training corpora to obtain at least two target corpora includes: A preset overlay algorithm is invoked to overlay the corpus to be determined with at least two expanded pre-training corpora. The result of the superposition process is subjected to sentence smoothing to obtain at least two target corpora, in which the probability of each word appearing is not 0.

4. A device for determining a customer profile, characterized in that, include: An acquisition unit is used to acquire pre-training corpora from a pre-training corpus, wherein the pre-training corpus contains at least two pre-training corpora. The acquisition unit is also used to acquire the corpus to be determined; The scaling unit is used to determine the target size of the at least two pre-training corpora and to scale the different pre-training corpora of the pre-training corpus according to the target size. The merging unit is used to merge the corpus to be determined with at least two scaled pre-training corpora to obtain at least two target corpora. A determining unit is configured to input the at least two target corpora into a unary recognition model, and within the unary recognition model determine the word frequency of each word in the corpus to be determined in each target corpus; The probability value of each sentence in the corpus to be determined is calculated based on the word frequency in each target corpus; Based on the probability value of each sentence, the target corpus corresponding to the maximum probability value is determined, and the customer profile identification information of the target corpus is obtained as the customer profile recognition result of the corpus to be determined. The unary recognition model is used to identify individual words in the corpus. The customer profile recognition results obtained from the unary recognition model outputting the corpus to be determined include: Wherein, the word frequency p( This can be achieved through the following formula: p( ) =count( ) / count(words) The probability value of each sentence is achieved using the following formula: p(s) = p( )p( )p( )...p( ) or, log p(s) = log p( )+log p( ) + log p( ) + ... + log p( ) in, This indicates the words or phrases appearing in the target corpus, where 's' represents sentences in the corpus to be determined; count( ) represents a word The count(words) represents the total number of times all words appear in the target corpus.

5. The apparatus according to claim 4, characterized in that, The telescopic unit includes: The first determining module is used to determine the scaling factor of each of the at least two pre-training corpora based on the target size; A scaling module is used to scale the data according to the scaling coefficients of the at least two pre-trained corpora.

6. The apparatus according to claim 4, characterized in that, The merging unit includes: The calling module is used to call a preset superposition algorithm to superimpose the corpus to be determined with at least two stretched pre-training corpora respectively; The processing module is used to perform sentence smoothing on the result of the superposition processing to obtain the at least two target corpora, wherein the probability of each word appearing in the at least two target corpora is not 0; The determining unit includes: The second determining module is used to determine the word frequency of each word in the corpus to be determined in the target corpus within the unary recognition model; The calculation module is used to calculate the probability value of each sentence in the corpus to be determined in each target corpus based on the word frequency; The third determining module is used to determine the customer profile recognition result based on the probability value of each sentence.

7. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-3.

8. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-3.

9. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method according to any one of claims 1-3.

Images