Enterprise supply and demand information acquisition method, system, equipment and medium

Abstract

The invention provides an enterprise supply and demand information acquisition method, system and device and a medium, and belongs to the field of natural language processing. The method comprises the following steps: accessing a preset enterprise website list and capturing supply and demand information by utilizing a web crawler technology; classifying and labeling the supply and demand information by adopting a pre-trained classification model and an attribute labeling model to obtain the category, the attribute and the label information of the supply and demand information; and based on the attribute and the label information, adopting a supply and demand matching algorithm to match potential supply and demand parties, and sending a matching information notification to potential supply and demand party enterprises. By capturing and updating enterprise supply and demand information in real time, the timeliness and accuracy of market information are ensured, so that the market failure phenomenon caused by information lag is reduced, potential supply and demand parties can be quickly found by using a supply and demand matching algorithm, the market efficiency is improved, an enterprise can know supply and demand changes on the market in time, and the market competitiveness is improved. Therefore, the production plan and the sales strategy can be quickly adjusted to adapt to the market demand.

Description

Technical Field

[0001] The present invention belongs to the technical field of natural language processing, and specifically relates to a method, system, device and medium for acquiring supply and demand information of an enterprise. Background Art

[0002] In the current market economy, the supply and demand relationship between enterprises is dynamic and complex, and the timeliness and accuracy of information are crucial to the decision-making of enterprises. Traditional ways of obtaining enterprise supply and demand information mainly rely on manual collection, industry reports, exhibition exchanges and other channels. These methods are not only time-consuming and labor-intensive, but also difficult to ensure the timeliness and comprehensiveness of information. With the popularization of the Internet, a large amount of enterprise supply and demand information has begun to be published on various websites, forums and social media, but this information is scattered, in various formats, and contains a lot of noise, making it difficult for enterprises to efficiently filter out valuable information.

[0003] In addition, due to information asymmetry and lag, there is often a mismatch between supply and demand in the market, resulting in resource waste and reduced market efficiency. For example, some companies may be in urgent need of specific raw materials or parts, but due to the inability to obtain supplier information in a timely manner, they have to face production stagnation or cost increases; at the same time, other companies may have excess production capacity or inventory, but due to the lack of effective demand information, they cannot convert these resources into revenue. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a method, system, device and medium for obtaining enterprise supply and demand information, which is used to fully or at least partially solve the technical problems of supply and demand mismatch and market inefficiency due to information asymmetry and lag in the above-mentioned prior art.

[0005] In a first aspect, an embodiment of the present application provides a method for obtaining supply and demand information of an enterprise, comprising: Using web crawler technology, access a preset list of corporate websites and crawl supply and demand information on the corporate websites; Using a pre-trained classification model and attribute labeling model to classify and label the supply and demand information, and obtain category information, attribute information and label information of the supply and demand information; Based on the attribute information and tag information of the supply and demand information, a supply and demand matching algorithm is adopted to match potential supply and demand parties, and a matching information notification is sent to the potential supply and demand party enterprises.

[0006] Optionally, use web crawler technology to access a preset list of corporate websites and crawl supply and demand information on the corporate websites, including: Design a crawler program using web crawler technology, wherein the crawler program uses a 5-second request interval and dynamic proxy IP technology to stably obtain data during long-term operation; Use crawlers to regularly visit a preset list of corporate websites to capture supply and demand information.

[0007] Optionally, a pre-trained classification model and attribute labeling model are used to classify and label the supply and demand information to obtain category information, attribute information and label information of the supply and demand information, including: De-duplication, de-noising, format unification and text cleaning are performed on the supply and demand information to obtain target supply and demand information data; Performing text segmentation and part-of-speech tagging on the target supply and demand information data, wherein the word segmentation tool Jieba is used to segment the continuous text into multiple independent words and phrases according to Chinese grammatical rules and context information, and marking the corresponding part of speech for each word and phrase; Train a classification model to identify supply and demand information categories: Some manually annotated data are used as training sets, wherein a category label is assigned to each text data during manual annotation; Select a naive Bayes model, and train the naive Bayes model with the labeled training set, continuously optimize the parameters of the naive Bayes model, and obtain a classification model; Train the annotation model: Using manually annotated attribute data as the training set, where a label is assigned to each attribute information in the text during annotation; Select a NER model and use the labeled training set to train the NER model to obtain a labeled model; After completing the classification and labeling of supply and demand information, the results are integrated and output.

[0008] Optionally, after completing the classification and labeling of supply and demand information, the results are integrated and output, including: The supply and demand information is spliced ​​according to the category labels and attribute information to form a structured data format, in which each data contains the category label and key attribute information; The integrated structured data format is verified and the verified data is input into the database for storage.

[0009] Optionally, based on the attribute information and tag information of the supply and demand information, a supply and demand matching algorithm is used to match potential supply and demand parties, including: The attribute information of the supply and demand information is used as a classification label, and the supply and demand information is used as the data to be labeled to obtain the training samples required for classification, which are expressed as x1=[x11, x12, x13, ..., x1n], and the corresponding result set y={L1, L2, ..., Lm}; Based on the training examples, classification learning is performed, and a matching model Y=f(x) is established. For unknown samples x, a prediction label is performed according to the matching model, and the holding probability is calculated, wherein the greater the holding probability, the higher the correlation between the label and the feature.

[0010] Optionally, the supply-demand matching algorithm selects a multi-label classification algorithm for training. The specific process is as follows: Prior probability estimate: The multi-label classification ML-KNN algorithm is used to calculate the prior probability of each label; The multi-label classification ML-KNN algorithm is used for training and the prior probability of each label is calculated; Finding k nearest neighbors: For each test sample in the test set, the multi-label classification ML-KNN algorithm is used to find the k nearest neighbors in the training set; Posterior probability estimate: For each test sample in the test set, the multi-label classification ML-KNN algorithm is used to calculate the posterior probability of each label based on the label information of the k nearest neighbors found; Determination of category labels: The category of the test sample is determined using the majority voting method or the weighted average method based on the posterior probabilities of the neighbors of each test sample in the test set.

[0011] Optionally, calculate the prior probability of each label according to the following formula:

[0012] In the formula, I(x i =t) represents the indicator function, which is 1 when the label set of sample xi contains label t, otherwise it is 0, n represents the total number of samples, and s represents the smoothing coefficient.

[0013] In a second aspect, the embodiment of the present application further provides a system for acquiring enterprise supply and demand information, including: The data collection layer is used to use web crawler technology to access a preset list of corporate websites and capture supply and demand information on the corporate websites; The data processing layer is used to classify and label the supply and demand information using a pre-trained classification model to obtain category information, attribute information and label information of the supply and demand information; The supply and demand matching layer is used to match potential supply and demand parties using a supply and demand matching algorithm based on the attribute information and tag information of the supply and demand information; The information push layer is used to send matching information notifications to potential supply and demand companies.

[0014] In a third aspect, an embodiment of the present application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for acquiring enterprise supply and demand information when executing the program.

[0015] In a fourth aspect, an embodiment of the present application further provides a storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method for obtaining the supply and demand information of enterprises described above are implemented.

[0016] It can be seen from the above technical solutions that the present invention has the following advantages: The method, system, device and medium for obtaining enterprise supply and demand information provided in this application ensure the timeliness and accuracy of market information by real-time capture and updating of enterprise supply and demand information, thereby reducing market failures caused by information lags. The supply and demand matching algorithm can quickly find potential supply and demand parties, improve market efficiency, and enterprises can promptly understand changes in supply and demand in the market, thereby quickly adjusting their production plans and sales strategies to adapt to market demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use in the description will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 A flowchart of a method for acquiring enterprise supply and demand information provided by an embodiment of the present invention; Figure 2 A detailed flow chart of a method for acquiring enterprise supply and demand information provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a system for acquiring supply and demand information of an enterprise provided by an embodiment of the present invention; Figure 4 A schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to better understand the technical solution of the present invention, the technical terms involved in the present invention are now explained: Web crawlers are programs or scripts that automatically crawl the World Wide Web according to certain rules. They are widely used in Internet search engines or other similar websites. They can automatically collect all the page contents they can access to obtain or update the content and retrieval methods of these websites. Web crawlers can easily obtain data from specified web pages. They can crawl the public data of websites, model and analyze the data, and generate data reports that are beneficial to themselves.

[0020] The TextRank algorithm is a graph-based sorting algorithm for text. Its basic idea comes from Google's PageRank algorithm. By dividing the text into several components (words, sentences) and building a graph model, the voting mechanism is used to sort the important components in the text. Only the information of a single document can be used to extract keywords and abstracts. Unlike models such as LDA and HMM, TextRank does not require prior learning and training of multiple documents. It is widely used because of its simplicity and effectiveness. The TextRank algorithm first performs a word segmentation operation on the provided sentence, and the obtained word segments are put into a set. The importance of the word segmentation mainly depends on the number of neighbors before and after the word segmentation. The more neighbors there are, the more votes there are for the word segmentation, the higher the weight, and the more important it is. The more the word segmentation appears, the more neighbors it has; the closer the word segmentation is in the middle (compared to the beginning and the end), the more neighbors it has. The main uses of the TextRank algorithm are two aspects. One is to extract the more important keywords in the text, and the other is to select which keyword appears more frequently in a paragraph.

[0021] The multi-label learning problem is a research hotspot in the field of international machine learning. It originally originated from the ambiguous problem encountered in document classification problems. In the traditional supervised learning framework, there is a one-to-one correspondence between real-world objects and their conceptual labels. It is generally believed that such learning problems are unambiguous. This type of problem is called a single-label classification problem, that is, a sample has only a single label. However, in real-world problems, ambiguous objects are widely present. Due to the existence of ambiguous problems, a sample may be associated with multiple labels. This type of problem is a multi-label classification problem. Multi-label learning has a wide range of applications in real life, such as automatic video annotation, bioinformatics, Web mining, information retrieval, personalized recommendation and other real-world applications.

[0022] In the detailed description below, various embodiments of the present disclosure will be described more fully. The present disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of the present disclosure to the specific embodiments disclosed herein, but rather the present disclosure should be understood to cover all adjustments, equivalents and / or alternatives that fall within the spirit and scope of the various embodiments of the present disclosure.

[0023] Hereinafter, the terms "include" or "may include" that may be used in various embodiments of the present disclosure indicate the presence of the disclosed functions or operations, and do not limit the addition of one or more functions or operations. In addition, as used in various embodiments of the present disclosure, the terms "include", "have" and their cognates are intended only to indicate specific features, numbers, steps, operations, or combinations of the foregoing items, and should not be understood as first excluding the presence of one or more other features, numbers, steps, operations, or combinations of the foregoing items or the possibility of adding one or more features, numbers, steps, operations, or combinations of the foregoing items.

[0024] In various embodiments of the present disclosure, the expression "or" or "at least one of A or / and B" includes any combination or all combinations of the words listed at the same time. For example, the expression "A or B" or "at least one of A or / and B" may include A, may include B, or may include both A and B.

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] See also Figure 1 The figure is a flow chart of a method for obtaining enterprise supply and demand information in a specific embodiment, including the following execution steps: Step 100: Using web crawler technology, access a preset list of corporate websites and crawl supply and demand information on the corporate websites.

[0027] Specifically, when executing step 100, the following steps may be specifically performed: S1000: Design a crawler program using web crawler technology, wherein the crawler program uses a 5-second request interval and dynamic proxy IP technology to stably obtain data during long-term operation.

[0028] S1001: Use crawler programs to regularly visit a preset list of corporate websites to capture supply and demand information.

[0029] Exemplarily, web crawler technology is used to design and implement an efficient and stable crawler program. The crawler program will regularly visit a preset list of corporate websites to capture supply and demand information on these websites. To ensure the comprehensiveness and accuracy of the data, the crawler program adopts a variety of strategies to deal with anti-crawler mechanisms, such as setting a reasonable request interval, using proxy IPs, and simulating user behavior. To ensure the comprehensiveness and accuracy of the data, the crawler program uses a 5-second request interval and dynamic proxy IP technology, which can still obtain data efficiently and stably during long-term operation. The collected raw data will be stored in a distributed database for subsequent processing and analysis. The database design will take into account the scalability and query efficiency of the data, and adopt appropriate indexing and partitioning strategies.

[0030] Step 101: Use a pre-trained classification model and attribute labeling model to classify and label the supply and demand information to obtain category information, attribute information and label information of the supply and demand information.

[0031] Specifically, when executing step 101, the following steps may be specifically performed: S1010: De-duplication, de-noising, format unification and text cleaning are performed on the supply and demand information to obtain target supply and demand information data.

[0032] For example, the data is first preprocessed, which mainly includes the following steps: (1) Deduplication: Delete duplicate data records to avoid repeated calculations in subsequent processing. (2) Denoising: Remove irrelevant information, such as advertisements and irrelevant text, to improve data quality. (3) Format unification: Unify data from different sources into the same format for subsequent processing. (4) Text cleaning: Further clean the text data to remove garbled characters and special characters to ensure the purity of the text data.

[0033] S1011: Performing text segmentation and part-of-speech tagging on the target supply and demand information data, wherein the word segmentation tool Jieba is used to segment the continuous text into multiple independent words and phrases according to Chinese grammatical rules and context information, and the corresponding part of speech is marked for each word and phrase.

[0034] For example, parts of speech include nouns, verbs, and adjectives.

[0035] S1012: Training a classification model to identify supply and demand information categories: Some manually annotated data are used as training sets, where each text data is assigned a category label during manual annotation: “supply”, “demand”; Select a naive Bayes model, and train the naive Bayes model with the labeled training set, continuously optimize the parameters of the naive Bayes model, and obtain a classification model; S1013: Training the annotation model: Using manually annotated attribute data as the training set, during annotation, a label is assigned to each attribute information in the text, such as "product name", "price", "specification", etc. Select a NER model and use the labeled training set to train the NER model to obtain a labeled model. Since the NER model may have misidentification or missed recognition, the extracted attribute information needs to be manually verified and corrected; S1014: After completing the classification and labeling of the supply and demand information, the results are integrated and output.

[0036] More specifically, when executing step S1014, the following steps may be specifically performed: splicing the supply and demand information according to the category labels and attribute information to form a structured data format, wherein each data contains the category labels and key attribute information; verifying the integrated structured data format, and inputting the verified data into the database for storage.

[0037] Through the above steps, the system can realize automatic classification and labeling of supply and demand information, and provide accurate and rich feature information for subsequent intelligent matching.

[0038] Step 102: Based on the attribute information and tag information of the supply and demand information, a supply and demand matching algorithm is used to match potential supply and demand parties, and a matching information notification is sent to the potential supply and demand party enterprises.

[0039] In a specific implementation, when executing step 102, the following steps may be specifically performed: using the attribute information of the supply and demand information as a classification label, and the supply and demand information as data to be labeled, obtaining training samples required for classification, and expressing them as x1=[x11, x12, x13, …, x1n], corresponding to a result set y={L1, L2, …, Lm} (the label L takes a value of 0 or 1, 0 indicates that the sample does not have the label, and 1 indicates that the sample has the label); performing classification learning based on the training samples, and establishing a matching model Y=f(x); for unknown samples x, predicting labels according to the matching model, and calculating the holding probability, matching potential supply and demand parties according to the holding probability, wherein the greater the holding probability, the higher the correlation between the label and the feature.

[0040] More specifically, the supply-demand matching algorithm selects a multi-label classification algorithm for training. The specific process is as follows: Prior probability estimate: The multi-label classification ML-KNN algorithm is used to calculate the prior probability of each label; The multi-label classification ML-KNN algorithm is used for training and the prior probability of each label is calculated; Finding k nearest neighbors: For each test sample in the test set, the multi-label classification ML-KNN algorithm is used to find the k nearest neighbors in the training set; Posterior probability estimate: For each test sample in the test set, the multi-label classification ML-KNN algorithm is used to calculate the posterior probability of each label based on the label information of the k nearest neighbors found; Determination of category labels: The category of the test sample is determined using the majority voting method or the weighted average method based on the posterior probabilities of the neighbors of each test sample in the test set.

[0041] Exemplarily, the prior probability of each label is calculated according to the following formula:

[0042] In the formula, I(x i =t) represents the indicator function, which is 1 when the label set of sample xi contains label t, otherwise it is 0, n represents the total number of samples, and s represents the smoothing coefficient.

[0043] In some implementations, in order to improve matching accuracy and efficiency, the matching strategy will be continuously optimized and adjusted. For example, by introducing a user feedback mechanism, the matching results can be evaluated and adjusted in real time.

[0044] In this embodiment, an efficient information push mechanism is designed and implemented. When the supply-demand matching algorithm finds potential supply and demand parties, the system will automatically send matching information notifications to these companies. The notification content will include key information such as basic information of the supply and demand parties, contact information, etc. To ensure the timeliness and effectiveness of information push, the system will use a variety of push methods, such as SMS, email, and in-site messages. At the same time, the system will also provide a user-defined push setting function, allowing companies to choose the appropriate push method and time according to their own needs.

[0045] By capturing and updating the supply and demand information of enterprises in real time, the timeliness and accuracy of market information are ensured, thereby reducing the market failure caused by information lag. The use of intelligent supply and demand matching algorithms can quickly and accurately find potential supply and demand parties, reduce transaction costs, and improve market efficiency. Enterprises can promptly understand the changes in supply and demand in the market, so as to quickly adjust their production plans and sales strategies to adapt to market demand. The pushed matching supply and demand information provides enterprises with new business opportunities and market space, which helps enterprises expand their business scope, increase market share, and thus enhance their market competitiveness. Moreover, through an efficient supply and demand matching mechanism, resources can flow to the demand side more accurately, avoiding waste and mismatch of resources. At the same time, enterprises can arrange production and sales activities more reasonably according to the information provided by the system, and realize the optimal allocation and utilization of resources.

[0046] In one embodiment, Figure 2 This is a detailed flow chart of a method for acquiring enterprise supply and demand information according to an embodiment of the present invention. This embodiment is further optimized and expanded on the basis of the above embodiments.

[0047] S200: Using web crawler technology, access a preset list of corporate websites and crawl supply and demand information on the corporate websites.

[0048] S201: De-duplication, de-noising, format unification and text cleaning are performed on the supply and demand information to obtain target supply and demand information data.

[0049] S202: Performing text segmentation and part-of-speech tagging on the target supply and demand information data, wherein the word segmentation tool Jieba is used to segment the continuous text into multiple independent words and phrases according to Chinese grammatical rules and context information, and the corresponding part of speech is marked for each word and phrase.

[0050] S203: using part of the manually annotated data as a training set, wherein during the manual annotation, a category label is assigned to each text data.

[0051] S204: Select a naive Bayes model, and train the naive Bayes model using the labeled training set, continuously optimize the parameters of the naive Bayes model, and obtain a classification model.

[0052] S205: Using manually annotated part of the attribute data as a training set, wherein during the annotation, a label is assigned to each attribute information in the text.

[0053] S206: Select a NER model, and use the labeled training set to train the NER model to obtain a labeled model.

[0054] S207: After completing the classification and labeling of the supply and demand information, the results are integrated and output.

[0055] S208: Use the attribute information of the supply and demand information as classification labels, and the supply and demand information as data to be labeled, obtain training samples required for classification, and express them as x1=[x11, x12, x13, ..., x1n], corresponding to the result set y={L1, L2, ..., Lm}.

[0056] S209: Classification learning is performed based on the training samples, and a matching model Y=f(x) is established. For unknown samples x, a prediction label is performed according to the matching model, and the holding probability is calculated. Potential supply and demand parties are matched according to the holding probability, wherein the greater the holding probability, the higher the correlation between the label and the feature.

[0057] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0058] like Figure 3 As shown, the following is an embodiment of the system for obtaining enterprise supply and demand information provided by the embodiment of the present disclosure, which belongs to the same inventive concept as the method for obtaining enterprise supply and demand information in the above-mentioned embodiments. For details not described in detail in the embodiment of the system for obtaining enterprise supply and demand information, please refer to the embodiment of the method for obtaining enterprise supply and demand information mentioned above.

[0059] The data collection layer is used to use web crawler technology to access a preset list of corporate websites and capture supply and demand information on the corporate websites; The data processing layer is used to classify and label the supply and demand information using a pre-trained classification model to obtain category information, attribute information and label information of the supply and demand information; The supply and demand matching layer is used to match potential supply and demand parties using a supply and demand matching algorithm based on the attribute information and tag information of the supply and demand information; The information push layer is used to send matching information notifications to potential supply and demand companies.

[0060] Through the wide application of this system, enterprises can more easily obtain the latest supply and demand information and technological trends in the industry, thereby promoting industrial upgrading and transformation. The market insight and data analysis functions provided by the system can also help enterprises discover new market trends and opportunities, and guide them to innovate and change.

[0061] Figure 4 It is a schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0062] The method for obtaining enterprise supply and demand information provided in the embodiment of the present application can be applied to electronic devices. It will be appreciated by those skilled in the art that the electronic device structure involved in the embodiment of the present invention does not constitute a limitation on the electronic device, and the electronic device may include more or less components than shown, or combine certain components, or arrange different components. In an embodiment of the present invention, the electronic device includes but is not limited to a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.

[0063] The electronic device may include a processor, an external memory interface, an internal memory, a universal serial bus (USB) interface, a charging management module, a power management module, a battery, a wireless communication module, an audio module, a speaker, a microphone, a sensor module, buttons, a camera, a display, and a SIM card interface, etc.

[0064] It is to be understood that the structure illustrated in the embodiments of the present application does not constitute a specific limitation on the electronic device. In other embodiments of the present application, the electronic device may include more or fewer components than shown in the figure, or combine certain components, or split certain components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0065] The processor may include one or more processing units, for example, the processor may include a central processing unit (CPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Among them, different processing units may be independent devices or integrated into one or more processors.

[0066] The processor can be the nerve center and command center of the electronic device. The controller can generate an operation control signal according to the instruction operation code and timing signal to complete the control of fetching and executing instructions.

[0067] A memory may also be provided in the processor for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. The memory may store instructions or data that the processor has just used or is cyclically used. If the processor needs to use the instruction or data again, it may be directly called from the memory. This avoids repeated access, reduces the waiting time of the processor, and thus improves system efficiency.

[0068] The external memory interface can be used to connect an external memory card, such as a MicroSD card, to expand the storage capacity of the electronic device. The external memory card communicates with the processor through the external memory interface to implement data storage functions. For example, files such as music and videos can be saved in the external memory card.

[0069] The internal memory can be used to store computer executable program codes, which include instructions. The processor executes various functional applications and data processing of the electronic device by running the instructions stored in the internal memory. The internal memory may include a program storage area and a data storage area. The internal memory may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc.

[0070] The wireless communication function of an electronic device can be realized through an antenna, a wireless communication module, a modem processor, and a baseband processor.

[0071] The wireless communication module can provide wireless communication solutions for electronic devices, including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared technology (IR), etc.

[0072] Electronic devices can implement audio functions, etc. through audio modules, speakers, receivers, microphones, headphone jacks, and application processors.

[0073] Electronic devices can achieve shooting functions through ISP, camera, video codec, GPU, display and application processor.

[0074] Electronic devices can achieve display functions through GPU, display screen and application processor.

[0075] The GPU is a microprocessor for image processing that connects the display screen and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. The processor may include one or more GPUs that execute program instructions to generate or change display information.

[0076] The display screen is used to display images, videos, etc. The display screen includes a display panel.

[0077] The storage medium provided in the present application stores a program product that can implement a method for acquiring enterprise supply and demand information.

[0078] The method for obtaining enterprise supply and demand information includes: using web crawler technology to access a preset list of enterprise websites and crawl the supply and demand information on the enterprise websites; using a pre-trained classification model and attribute labeling model to classify and label the supply and demand information, and obtain the category information, attribute information and label information of the supply and demand information; based on the attribute information and label information of the supply and demand information, using a supply and demand matching algorithm to match potential supply and demand parties, and sending matching information notifications to potential supply and demand party companies.

[0079] In some possible implementations, the subject name of the present disclosure, the method and system for obtaining enterprise supply and demand information can be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps described in the above "Exemplary Method" section of this specification according to various exemplary implementations of the present disclosure.

[0080] The storage medium of the present disclosure can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0081] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for obtaining enterprise supply and demand information, characterized in that: include: Using web crawler technology, access a preset list of corporate websites and crawl supply and demand information on the corporate websites; Using a pre-trained classification model and attribute labeling model to classify and label the supply and demand information, and obtain category information, attribute information and label information of the supply and demand information; Based on the attribute information and tag information of the supply and demand information, a supply and demand matching algorithm is adopted to match potential supply and demand parties, and a matching information notification is sent to the potential supply and demand party enterprises.

2. The method for obtaining enterprise supply and demand information according to claim 1, characterized in that: Using web crawler technology, we visit a preset list of corporate websites and crawl the supply and demand information on the corporate websites, including: Design a crawler program using web crawler technology, wherein the crawler program uses a 5-second request interval and dynamic proxy IP technology to stably obtain data during long-term operation; Use crawlers to regularly visit a preset list of corporate websites to capture supply and demand information.

3. The method for obtaining enterprise supply and demand information according to claim 1, characterized in that: Using a pre-trained classification model and attribute labeling model, the supply and demand information is classified and labeled to obtain category information, attribute information and label information of the supply and demand information, including: De-duplication, de-noising, format unification and text cleaning are performed on the supply and demand information to obtain target supply and demand information data; Performing text segmentation and part-of-speech tagging on the target supply and demand information data, wherein the word segmentation tool Jieba is used to segment the continuous text into multiple independent words and phrases according to Chinese grammatical rules and context information, and marking the corresponding part of speech for each word and phrase; Train a classification model to identify supply and demand information categories: Some manually annotated data are used as training sets, wherein a category label is assigned to each text data during manual annotation; Select a naive Bayes model, and train the naive Bayes model with the labeled training set, continuously optimize the parameters of the naive Bayes model, and obtain a classification model; Train the annotation model: Using manually annotated attribute data as the training set, where a label is assigned to each attribute information in the text during annotation; Select a NER model and use the labeled training set to train the NER model to obtain a labeled model; After completing the classification and labeling of supply and demand information, the results are integrated and output.

4. The method for obtaining enterprise supply and demand information according to claim 3, characterized in that: After completing the classification and labeling of supply and demand information, the results are integrated and output, including: The supply and demand information is spliced ​​according to the category labels and attribute information to form a structured data format, in which each data contains the category label and key attribute information; The integrated structured data format is verified and the verified data is input into the database for storage.

5. The method for obtaining enterprise supply and demand information according to claim 1, characterized in that: Based on the attribute information and tag information of the supply and demand information, a supply and demand matching algorithm is used to match potential supply and demand parties, including: The attribute information of the supply and demand information is used as a classification label, and the supply and demand information is used as the data to be labeled to obtain the training samples required for classification, which are expressed as x1=[x11, x12, x13, ..., x1n], and the corresponding result set y={L1, L2, ..., Lm}; Based on the training examples, classification learning is performed and a matching model Y=f(x) is established. For unknown samples x, prediction labels are performed according to the matching model, and the holding probability is calculated. Potential supply and demand parties are matched according to the holding probability. The greater the holding probability, the higher the correlation between the label and the feature.

6. The method for obtaining enterprise supply and demand information according to claim 5, characterized in that: The supply and demand matching algorithm selects the multi-label classification algorithm for training. The specific process is as follows: Prior probability estimate: The multi-label classification ML-KNN algorithm is used to calculate the prior probability of each label; The multi-label classification ML-KNN algorithm is used for training and the prior probability of each label is calculated; Finding k nearest neighbors: For each test sample in the test set, the multi-label classification ML-KNN algorithm is used to find the k nearest neighbors in the training set; Posterior probability estimate: For each test sample in the test set, the multi-label classification ML-KNN algorithm is used to calculate the posterior probability of each label based on the label information of the k nearest neighbors found; Determination of category labels: The category of the test sample is determined using the majority voting method or the weighted average method based on the posterior probabilities of the neighbors of each test sample in the test set.

7. The method for obtaining enterprise supply and demand information according to claim 6, characterized in that: The prior probability of each label is calculated according to the following formula: In the formula, I(x i =t) represents the indicator function, which is 1 when the label set of sample xi contains label t, otherwise it is 0, n represents the total number of samples, and s represents the smoothing coefficient.

8. A system for acquiring enterprise supply and demand information, characterized in that: include: The data collection layer is used to use web crawler technology to access a preset list of corporate websites and capture supply and demand information on the corporate websites; The data processing layer is used to classify and label the supply and demand information using a pre-trained classification model to obtain category information, attribute information and label information of the supply and demand information; The supply and demand matching layer is used to match potential supply and demand parties using a supply and demand matching algorithm based on the attribute information and tag information of the supply and demand information; The information push layer is used to send matching information notifications to potential supply and demand companies.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method for obtaining enterprise supply and demand information as described in any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for obtaining enterprise supply and demand information as described in any one of claims 1 to 7 are implemented.

Images