Address fuzzy matching method, device and equipment and readable storage medium

Abstract

The invention discloses an address fuzzy matching method, device and equipment and a readable storage medium, through extracting address elements of a target address text, constructing an address map and utilizing a map attention network model to extract an address feature vector fused with address semantics and structure information, a non-standard address can be better dealt with, and meanwhile, the address fuzzy matching efficiency is improved. The vector database is applied to perform matching search of neighbor address vectors, the problem of time consumption does not need to be considered, and finally an expected fuzzy matching address is screened out according to a preset threshold value, so that the problems that an existing address matching method depends on a normative address database and a manual exhaustion address structure, is difficult to cope with non-standard addresses and cannot cope with other addresses are solved. And when large-scale data processing is carried out, the time consumed by pair-by-pair calculation is long, and the method is difficult to adapt to all scenes.

Description

Technical Field

[0001] The present application relates to the field of address matching technology, and in particular to an address fuzzy matching method, device, equipment and readable storage medium. Background Art

[0002] Driven by digital technology, the financial industry is undergoing unprecedented changes. The expansion of online users has brought new growth opportunities to financial institutions, but it has also brought new challenges and risks, especially the specialization of fraudulent behavior and the evolution of methods. Gang fraud has become an important risk point in financial anti-fraud due to its strict organization and strong concealment. By analyzing the correlation between customer address information, a customer association model can be constructed to assist in identifying fraud gangs.

[0003] In the financial credit scenario, modeling and fuzzy matching of customer address information is an effective means. At present, the mainstream address matching methods are mainly divided into address matching based on structural analysis and address matching based on semantic representation.

[0004] Among them, the structural analysis method extracts address elements, analyzes the syntactic pattern of address description, constructs an address structure tree, and then enumerates the rule structure and performs matching using a dictionary. However, the structural analysis method relies on a standardized address database and manually enumerates the address structure, but it is difficult to deal with non-standard addresses, resulting in limited matching accuracy.

[0005] Address matching based on semantic representation transforms address semantic expression and matching algorithm into deep learning modeling, captures text semantic features through training and optimization of address language model, and supports address matching. Although this method improves accuracy through deep learning, it takes a long time to calculate pair by pair when facing large-scale data processing, and it is difficult to adapt to all scenarios. Summary of the invention

[0006] The present application provides an address fuzzy matching method, apparatus, device and readable storage medium, which solves the technical problems of existing address matching methods, such as reliance on standardized address databases and manually exhaustive address structures, difficulty in dealing with non-standard addresses, resulting in limited matching accuracy, and time-consuming pair-by-pair calculations when facing large-scale data processing, making it difficult to adapt to all scenarios.

[0007] In view of this, the first aspect of the present application provides an address fuzzy matching method, the method comprising:

[0008] Get the target address text to be matched;

[0009] After preprocessing the target address text, the recalled address is screened and matched based on a preset POI database and a preset rule template;

[0010] If the screening match fails, the target address text is parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing results;

[0011] Based on a preset graph attention network model and in combination with the address graph, vector encoding and feature extraction are performed on the target address text to obtain an address feature vector of the target address text;

[0012] Searching for at least one neighboring address vector of the address feature vector in a vector database, and calculating a matching degree between the neighboring address vector and the address feature vector based on a preset matching model;

[0013] If the matching degree is higher than a preset threshold, the address information corresponding to the neighbor address vector is used as the fuzzy matching result of the target address text matching.

[0014] Optionally, after preprocessing the target address text, screening and matching the recalled address based on a preset POI database and a preset rule template includes:

[0015] Cleaning the target address text and preprocessing it to a uniform format;

[0016] Matching the preprocessed target address text with a preset POI database to obtain a number of matching recall addresses;

[0017] It is determined whether the target address text and the matched recalled addresses have the same POI element or the same address classification information.

[0018] Optionally, after preprocessing the target address text, screening and matching of the recalled addresses are performed based on a preset POI database and a preset rule template, and then the method further includes:

[0019] If the target address text and the matched recall addresses have the same POI element or the same address classification information, the matched recall addresses are element parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing results.

[0020] Optionally, if the screening match fails, performing element parsing on the target address text based on the entity recognition model, and constructing an address graph according to the address element parsing result, including:

[0021] If there is no recall address with the same POI element or the same address classification information among the target address text and the matched plurality of the recall addresses, then based on the Bert model pre-trained in the address entity corpus, the target address text is element parsed to obtain the address element parsing result of the target address text;

[0022] According to the parsing result of the address element, an address graph is constructed with the address element as a parent node to determine the associated path of the target address text.

[0023] Optionally, the preset graph attention network model is combined with the address graph to perform vector encoding and feature extraction on the target address text to obtain an address feature vector of the target address text, including:

[0024] After converting the address elements contained in the association path into a computable format, random sampling is performed based on the node2vec random walk strategy, starting from each node in the association path to obtain a node sequence, and the address embedding vector corresponding to the node sequence is generated through the skip-gram model;

[0025] The address embedding vector is used as the input of a preset graph attention network model, and the preset graph attention network model outputs an address feature vector, which integrates address semantics and address hierarchical structure information.

[0026] Optionally, searching for at least one neighboring address vector of the address feature vector in a vector database, and calculating a matching degree between the neighboring address vector and the address feature vector based on a preset matching model, specifically includes:

[0027] Using IVF-FLAT method to search for the nearest address vector of at least one address feature vector in the vector database;

[0028] The matching degree between the neighbor address vector and the address feature vector is calculated based on a preset matching model.

[0029] Optionally, the preset matching model is specifically:

[0030] respectively calculating a similarity score between the address feature vector and the neighboring address vector, and a coincidence score between the address feature vector and the neighboring address vector;

[0031] The similarity score and the coincidence score are summed based on a preset reconciliation coefficient to obtain a matching degree between the neighbor address vector and the address feature vector.

[0032] A second aspect of the present application provides an address fuzzy matching device, the device comprising:

[0033] An acquisition unit, used for acquiring a target address text to be matched;

[0034] A first matching unit, configured to pre-process the target address text and then perform screening and matching of the recalled addresses based on a preset POI database and a preset rule template;

[0035] A first processing unit, configured to perform element parsing on the target address text based on an entity recognition model if the screening match fails, and construct an address graph according to the address element parsing result;

[0036] A feature extraction unit, configured to perform vector encoding and feature extraction on the target address text based on a preset graph attention network model and in combination with the address graph, so as to obtain an address feature vector of the target address text;

[0037] A second matching unit, configured to search a vector database for a neighboring address vector of at least one of the address feature vectors, and calculate a matching degree between the neighboring address vector and the address feature vector based on a preset matching model;

[0038] The second processing unit is configured to use the address information corresponding to the neighboring address vector as a fuzzy matching result of the target address text matching if the matching degree is higher than a preset threshold.

[0039] Optionally, the first matching unit is specifically configured to:

[0040] Cleaning the target address text and preprocessing it to a uniform format;

[0041] Matching the preprocessed target address text with a preset POI database to obtain a number of matching recall addresses;

[0042] It is determined whether the target address text and the matched recalled addresses have the same POI element or the same address classification information.

[0043] Optionally, it also includes:

[0044] The third processing unit is used to perform element parsing on the matched recall addresses based on the entity recognition model if there are recall addresses with the same POI element or the same address classification information among the target address text and the matched recall addresses, and to construct an address graph according to the address element parsing results.

[0045] Optionally, the first processing unit is specifically configured to:

[0046] If there is no recall address with the same POI element or the same address classification information among the target address text and the matched plurality of the recall addresses, then based on the Bert model pre-trained in the address entity corpus, the target address text is element parsed to obtain the address element parsing result of the target address text;

[0047] According to the parsing result of the address element, an address graph is constructed with the address element as a parent node to determine the associated path of the target address text.

[0048] Optionally, the feature extraction unit is specifically used for:

[0049] After converting the address elements contained in the association path into a computable format, random sampling is performed based on the node2vec random walk strategy, starting from each node in the association path to obtain a node sequence, and the address embedding vector corresponding to the node sequence is generated through the skip-gram model;

[0050] The address embedding vector is used as the input of a preset graph attention network model, and the preset graph attention network model outputs an address feature vector, which integrates address semantics and address hierarchical structure information.

[0051] Optionally, the feature extraction unit is further used for:

[0052] The address feature vector is stored in a vector database.

[0053] Optionally, the second matching unit is specifically configured to:

[0054] Using the IVF-FLAT method, searching for a neighboring address vector of at least one of the address feature vectors in a vector database;

[0055] The matching degree between the neighbor address vector and the address feature vector is calculated based on a preset matching model, wherein the preset matching model is specifically:

[0056]

[0057] Among them, r1 and r2 are preset harmonic coefficients, represents the similarity score between the address feature vector and the neighboring address vector, It represents the overlap score between the address feature vector and the neighboring address vector, G is the calculation function of the number of address pair element overlaps, L is the calculation function of the length of the address element corresponding to the address feature vector, and k is the type of preset key address element.

[0058] A third aspect of the present application provides an address fuzzy matching device, the device comprising a processor and a memory:

[0059] The memory is used to store program code and transmit the program code to the processor;

[0060] The processor is used to execute the steps of the address fuzzy matching method as described in the first aspect according to the instructions in the program code.

[0061] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program code, and the program code is used to execute the method described in the first aspect above.

[0062] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0063] In the present application, a method, apparatus, device and readable storage medium for address fuzzy matching are provided. By extracting the address elements of the target address text, an address graph is constructed, and a graph attention network model is used to extract an address feature vector that integrates the address semantics and structural information. This method can better cope with non-standard addresses. At the same time, a vector database is used to perform a matching search of neighboring address vectors without considering time-consuming issues. Finally, the desired fuzzy matching address is screened out according to a preset threshold. This solves the technical problems of existing address matching methods, such as reliance on a standardized address database and manually exhaustive address structures, difficulty in coping with non-standard addresses, and limited matching accuracy. In addition, when faced with large-scale data processing, pair-by-pair calculations are time-consuming and difficult to adapt to all scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a method flow chart of the address fuzzy matching method in the embodiment of the present application;

[0065] Figure 2 This is a schematic diagram of the structure of the address fuzzy matching device in the embodiment of the present application;

[0066] Figure 3 It is a structural diagram of the address fuzzy matching device in the embodiment of the present application. DETAILED DESCRIPTION

[0067] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0068] The present application designs an address fuzzy matching method, device, equipment and readable storage medium, which solves the technical problems of existing address matching methods, such as reliance on standardized address databases and manually exhaustive address structures, difficulty in dealing with non-standard addresses, resulting in limited matching accuracy, and time-consuming pair-by-pair calculations when facing large-scale data processing, making it difficult to adapt to all scenarios.

[0069] For easier understanding, see Figure 1 , Figure 1 : is a method flow chart of the address fuzzy matching method in the embodiment of the present application, such as Figure 1 As shown, specifically:

[0070] Step S1, obtaining the target address text to be matched;

[0071] Step S2: After preprocessing the target address text, the recalled address is screened and matched based on a preset POI database and a preset rule template;

[0072] Step S3: If the screening match fails, the target address text is parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing results;

[0073] Step S4: Based on the preset graph attention network model and in combination with the address graph, vector encoding and feature extraction are performed on the target address text to obtain an address feature vector of the target address text;

[0074] Step S5, searching for a neighboring address vector of at least one address feature vector in a vector database, and calculating a matching degree between the neighboring address vector and the address feature vector based on a preset matching model;

[0075] Step S6: If the matching degree is higher than a preset threshold, the address information corresponding to the neighbor address vector is used as the fuzzy matching result of the target address text matching.

[0076] Furthermore, step S2 specifically includes:

[0077] Step S21, pre-processing the target address text for cleaning and format unification;

[0078] It should be noted that the address text is cleaned to remove useless symbols, spaces, etc.

[0079] Standardize the format, such as converting all texts to the same language or encoding format.

[0080] Step S22: matching the preprocessed target address text with a preset POI database to obtain a number of matching recall addresses;

[0081] It should be noted that the pre-processed address text is used to match the POI database. The POI (Points of Interest) information point database is a spatial database containing geographic location information and attribute information, and is mainly used to provide geographic location information for maps.

[0082] Filter the list of possible recall addresses.

[0083] Step S23: determine whether the target address text and the matched recalled addresses have the same POI element or the same address classification information.

[0084] It should be noted that the target address text and the recalled address are analyzed to see whether there are the same POI elements or address classification information.

[0085] Determine if further matching processing is required.

[0086] Furthermore, after step S2, the following steps are further included:

[0087] If the target address text and the matched recalled addresses have the same POI element or the same address classification information, the matched recalled addresses are parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing results.

[0088] It should be noted that the entity recognition model is used to perform element parsing on the recalled address.

[0089] Construct an address map for subsequent feature extraction.

[0090] Furthermore, step S3 specifically includes:

[0091] If the target address text and the matched recalled addresses do not have the same POI element or the same address classification information, then based on the Bert model pre-trained in the address entity corpus, the target address text is element parsed to obtain the address element parsing result of the target address text;

[0092] According to the parsing results of the address element, the address element is used as the parent node to build an address graph and determine the associated path of the target address text.

[0093] It should be noted that the pre-trained Bert model is used to perform element parsing on the target address text.

[0094] Build an address graph based on the parsing results and determine the associated path.

[0095] Since the address syntax has a hierarchical structure, and the order of the address elements after parsing naturally expresses this hierarchical relationship, this solution uses these relationships to construct the associated edges between address elements. We perform the above operations on each piece of data in the historical address database. Each address forms a path after being associated with the upper and lower hierarchical structures, and all address paths eventually form a huge undirected address graph.

[0096] The construction of the address map requires the collection of a large amount of historical address data, which is part of the modeling phase. When the address map is built, in actual business use, this phase only requires generating the associated path of the current address.

[0097] Furthermore, step S4 specifically includes:

[0098] After converting the address elements contained in the association path into a computable format, based on the node2vec random walk strategy, random sampling is performed with each node in the association path as the starting point to obtain a node sequence, and the address embedding vector corresponding to the node sequence is generated through the skip-gram model;

[0099] The address embedding vector is used as the input of the preset graph attention network model, and the preset graph attention network model outputs the address feature vector, which integrates the address semantics and address hierarchical structure information.

[0100] It should be noted that the address elements are converted into a computable format.

[0101] Use the node2vec strategy to perform random walks and generate node sequences.

[0102] Generate address embedding vectors through the skip-gram model.

[0103] Input the embedding vector into the graph attention network model to extract the address feature vector that integrates semantics and hierarchical structure.

[0104] The graph attention network model adopts a dual-tower structure, and the left and right substructure model parameters are shared. The model usage needs to be further explained. In the model training stage, this solution uses a large number of address pair data sets (data B) accumulated by the business as input. The data set format is {(address A, address B, label)}, for example, {("No. 316 Laoshan Road, Pudong New District, Shanghai", "Ping An Technology Business Building, No. 316 Laoshan Road, Lujiazui Street, Pudong New District, Shanghai", matched), ("Ping An Technology Business Building, No. 316 Laoshan Road, Pudong New District, Shanghai", "Ping An Building, No. 206 Kaibin Road, Xuhui District, Shanghai", not matched)...}, the model performs supervised training and outputs the feature vector and similarity score of the address pair; in the model reasoning stage, that is, the actual business application stage, the input of the model is a single address, and the model output is the feature vector of the address.

[0105] Furthermore, step S4 further includes:

[0106] The address feature vector is stored in the vector database.

[0107] Furthermore, step S5 specifically includes:

[0108] Using IVF-FLAT method to search for the nearest address vector of at least one address feature vector in the vector database;

[0109] The matching degree between the neighbor address vector and the address feature vector is calculated based on the preset matching model. The preset matching model is specifically:

[0110]

[0111] Among them, r1 and r2 are preset harmonic coefficients, represents the similarity score between the address feature vector and the neighboring address vector, It represents the overlap score between the address feature vector and its neighboring address vector. G is the function for calculating the number of overlaps between address pair elements. L is the function for calculating the length of address elements corresponding to the address feature vector. k is the type of preset key address element.

[0112] It should be noted that the IVF-FLAT method is used to search for neighbor address vectors in the vector database.

[0113] The preset matching model is used to calculate the matching degree between the target address feature vector and the neighboring address vector.

[0114] The matching model takes into account both similarity and overlap scores.

[0115] First, search the vector database to obtain the address vector set Vk = {v1, v2, v3, ...} that is the K nearest neighbors of the feature vector vc in space. Use the IVF_FLAT method when searching for vectors, divide the vectors in the vector database into nlist units, and calculate the distance between the target vector and the center point of these nlist units to obtain the nearest nprobe units, and then calculate the closest vector in the unit. This method avoids calculating the distance between the target vector and each vector in the vector database, further improving the search efficiency.

[0116] The current address vector vc and the neighboring address vector vi are sequentially sent to the trained matching model to obtain the matching score.

[0117] A pair of address vectors may be close in the vector space due to the large number of overlapping address elements, but in actual scenarios, the matching of "key" address key elements will generate greater business value. For example, in actual business scenarios, the matching value of addresses containing the same POI element will be higher than that of addresses containing Province, City, District, and Town elements. A typical example is that we prefer to get the match of the address pair "No. 316, Laoshan Road, Lujiazui Street, Pudong New District, Shanghai" and "Ping An Technology Business Building, Pudong New District, Shanghai" rather than the match of "No. 316, Laoshan Road, Lujiazui Street, Pudong New District, Shanghai" and "No. 500, Laoshan Road, Lujiazui Street, Pudong New District, Shanghai". Therefore, this solution adds the above considerations in the matching scoring process, and adds the score of the overlap of the "key" address elements of the address pair in addition to calculating the similarity score of the address pair vector.

[0118] See also Figure 2 , Figure 2 Schematic diagram of the structure of the address fuzzy matching device in the embodiment of the present application. Figure 2 As shown, specifically:

[0119] An acquisition unit 201 is used to acquire a target address text to be matched;

[0120] A first matching unit 202 is used to pre-process the target address text and then perform screening and matching of the recalled addresses based on a preset POI database and a preset rule template;

[0121] The first processing unit 203 is used to perform element parsing on the target address text based on the entity recognition model if the screening match fails, and to construct an address graph according to the address element parsing result;

[0122] The feature extraction unit 204 is used to perform vector encoding and feature extraction on the target address text based on a preset graph attention network model and in combination with the address graph to obtain an address feature vector of the target address text;

[0123] A second matching unit 205, configured to search for a neighboring address vector of at least one address feature vector in the vector database, and calculate a matching degree between the neighboring address vector and the address feature vector based on a preset matching model;

[0124] The second processing unit 206 is configured to use the address information corresponding to the neighboring address vector as the fuzzy matching result of the target address text matching if the matching degree is higher than a preset threshold.

[0125] Furthermore, the first matching unit 202 is specifically configured to:

[0126] Clean the target address text and pre-process it to unify the format;

[0127] Match the preprocessed target address text with the preset POI database to obtain a number of matching recall addresses;

[0128] It is determined whether the target address text and the matched recalled addresses have the same POI element or the same address classification information.

[0129] Furthermore, it also includes:

[0130] The third processing unit is used to perform element parsing on the matched recall addresses based on the entity recognition model if there are recall addresses with the same POI elements or the same address classification information among the target address text and the matched recall addresses, and to construct an address graph according to the address element parsing results.

[0131] Furthermore, the first processing unit 203 is specifically configured to:

[0132] If the target address text and the matched recalled addresses do not have the same POI element or the same address classification information, then based on the Bert model pre-trained in the address entity corpus, the target address text is element parsed to obtain the address element parsing result of the target address text;

[0133] According to the parsing results of the address element, the address element is used as the parent node to build an address graph and determine the associated path of the target address text.

[0134] Furthermore, the feature extraction unit 204 is specifically used for:

[0135] After converting the address elements contained in the association path into a computable format, based on the node2vec random walk strategy, random sampling is performed with each node in the association path as the starting point to obtain a node sequence, and the address embedding vector corresponding to the node sequence is generated through the skip-gram model;

[0136] The address embedding vector is used as the input of the preset graph attention network model, and the preset graph attention network model outputs the address feature vector, which integrates the address semantics and address hierarchical structure information.

[0137] Furthermore, the feature extraction unit 204 is also used for:

[0138] The address feature vector is stored in the vector database.

[0139] Furthermore, the second matching unit 205 is specifically configured to:

[0140] Using IVF-FLAT method to search for the nearest address vector of at least one address feature vector in the vector database;

[0141] The matching degree between the neighbor address vector and the address feature vector is calculated based on the preset matching model. The preset matching model is specifically:

[0142]

[0143] Among them, r1 and r2 are preset harmonic coefficients, represents the similarity score between the address feature vector and the neighboring address vector, It represents the overlap score between the address feature vector and its neighboring address vector. G is the function for calculating the number of overlaps between address pair elements. L is the function for calculating the length of address elements corresponding to the address feature vector. k is the type of preset key address element.

[0144] The present application also provides another address fuzzy matching device, such as Figure 3 For the sake of convenience, only the parts related to the embodiments of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiments of the present application. The terminal can be any terminal device including a mobile phone, a tablet computer, a personal digital assistant (English full name: Personal Digital Assistant, English abbreviation: PDA), a sales terminal (English full name: Point of Sales, English abbreviation: POS), a car computer, etc., taking the terminal as a mobile phone as an example:

[0145] Figure 3 FIG. 1 is a block diagram showing a partial structure of a mobile phone related to a terminal provided in an embodiment of the present application. Figure 3 The mobile phone includes: a radio frequency (RF) circuit 1010, a memory 1020, an input unit 1030, a display unit 1040, a sensor 1050, an audio circuit 1060, a wireless fidelity (WiFi) module 1070, a processor 1080, and a power supply 1090. Those skilled in the art can understand that Figure 3 The mobile phone structure shown in the figure does not constitute a limitation on the mobile phone, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0146] Combine the following Figure 3 A detailed introduction to the various components of the mobile phone:

[0147] The RF circuit 1010 can be used for receiving and sending signals during information transmission or calls. In particular, after receiving the downlink information of the base station, it is sent to the processor 1080 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit 1010 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (full name: Low Noise Amplifier, English abbreviation: LNA), a duplexer, etc. In addition, the RF circuit 1010 can also communicate with the network and other devices through wireless communication. The above-mentioned wireless communications can use any communication standard or protocol, including but not limited to Global System of Mobile communications (Global System of Mobile communication, English abbreviation: GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, English abbreviation: CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (Long Term Evolution, English abbreviation: LTE), e-mail, Short Messaging Service (SMS), etc.

[0148] The memory 1020 can be used to store software programs and modules. The processor 1080 executes various functional applications and data processing of the mobile phone by running the software programs and modules stored in the memory 1020. The memory 1020 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory 1020 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0149] The input unit 1030 can be used to receive input digital or character information, and to generate key signal input related to the user settings and function control of the mobile phone. Specifically, the input unit 1030 may include a touch panel 1031 and other input devices 1032. The touch panel 1031, also known as a touch screen, can collect the user's touch operation on or near it (such as the user's operation on the touch panel 1031 or near the touch panel 1031 using any suitable object or accessory such as a finger, stylus, etc.), and drive the corresponding connection device according to a pre-set program. Optionally, the touch panel 1031 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it to the processor 1080, and can receive and execute commands sent by the processor 1080. In addition, the touch panel 1031 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic waves. In addition to the touch panel 1031, the input unit 1030 may further include other input devices 1032. Specifically, the other input devices 1032 may include but are not limited to one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and the like.

[0150] The display unit 1040 can be used to display information input by the user or information provided to the user and various menus of the mobile phone. The display unit 1040 may include a display panel 1041. Optionally, the display panel 1041 may be configured in the form of a liquid crystal display (full name in English: Liquid Crystal Display, English abbreviation: LCD), an organic light-emitting diode (full name in English: Organic Light-Emitting Diode, English abbreviation: OLED), etc. Further, the touch panel 1031 may cover the display panel 1041. When the touch panel 1031 detects a touch operation on or near it, it is transmitted to the processor 1080 to determine the type of touch event. Subsequently, the processor 1080 provides a corresponding visual output on the display panel 1041 according to the type of touch event. Although in Figure 3 In the embodiment, the touch panel 1031 and the display panel 1041 are used as two independent components to realize the input and output functions of the mobile phone, but in some embodiments, the touch panel 1031 and the display panel 1041 can be integrated to realize the input and output functions of the mobile phone.

[0151] The mobile phone may also include at least one sensor 1050, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 1041 according to the brightness of the ambient light, and the proximity sensor may turn off the display panel 1041 and / or the backlight when the mobile phone is moved to the ear. As a type of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in all directions (generally three axes), and can detect the magnitude and direction of gravity when stationary. It can be used for applications that identify the posture of the mobile phone (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; as for other sensors that can be configured in the mobile phone, such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc., they will not be repeated here.

[0152] The audio circuit 1060, the speaker 1061, and the microphone 1062 can provide an audio interface between the user and the mobile phone. The audio circuit 1060 can transmit the received audio data to the speaker 1061 after converting the received audio data into an electrical signal, which is converted into a sound signal for output; on the other hand, the microphone 1062 converts the collected sound signal into an electrical signal, which is received by the audio circuit 1060 and converted into audio data, and then the audio data is output to the processor 1080 for processing, and then sent to another mobile phone through the RF circuit 1010, or the audio data is output to the memory 1020 for further processing.

[0153] WiFi is a short-range wireless transmission technology. The mobile phone can help users send and receive emails, browse web pages and access streaming media through the WiFi module 1070. It provides users with wireless broadband Internet access. Figure 3 A WiFi module 1070 is shown, but it is understandable that it is not an essential component of the mobile phone and can be omitted as needed without changing the essence of the invention.

[0154] The processor 1080 is the control center of the mobile phone. It uses various interfaces and lines to connect various parts of the entire mobile phone. By running or executing software programs and / or modules stored in the memory 1020 and calling data stored in the memory 1020, it executes various functions of the mobile phone and processes data, thereby monitoring the mobile phone as a whole. Optionally, the processor 1080 may include one or more processing units; preferably, the processor 1080 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 1080.

[0155] The mobile phone also includes a power supply 1090 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 1080 through a power management system, so that the power management system can manage functions such as charging, discharging, and power consumption.

[0156] Although not shown, the mobile phone may also include a camera, a Bluetooth module, etc., which will not be described in detail here.

[0157] In the embodiment of the present application, the processor 1080 included in the terminal also has the following functions:

[0158] Step S1, obtaining the target address text to be matched;

[0159] Step S2: After preprocessing the target address text, the recalled address is screened and matched based on a preset POI database and a preset rule template;

[0160] Step S3: If the screening match fails, the target address text is parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing results;

[0161] Step S4: Based on the preset graph attention network model and in combination with the address graph, vector encoding and feature extraction are performed on the target address text to obtain an address feature vector of the target address text;

[0162] Step S5, searching for a neighboring address vector of at least one address feature vector in a vector database, and calculating a matching degree between the neighboring address vector and the address feature vector based on a preset matching model;

[0163] Step S6: If the matching degree is higher than a preset threshold, the address information corresponding to the neighbor address vector is used as the fuzzy matching result of the target address text matching.

[0164] Furthermore, step S2 specifically includes:

[0165] Step S21, pre-processing the target address text for cleaning and format unification;

[0166] Step S22: matching the preprocessed target address text with a preset POI database to obtain a number of matching recall addresses;

[0167] Step S23: determine whether the target address text and the matched recalled addresses have the same POI element or the same address classification information.

[0168] Furthermore, after step S2, the following steps are further included:

[0169] If the target address text and the matched recalled addresses have the same POI element or the same address classification information, the matched recalled addresses are parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing results.

[0170] Furthermore, step S3 specifically includes:

[0171] If the target address text and the matched recalled addresses do not have the same POI element or the same address classification information, then based on the Bert model pre-trained in the address entity corpus, the target address text is element parsed to obtain the address element parsing result of the target address text;

[0172] According to the parsing results of the address element, the address element is used as the parent node to build an address graph and determine the associated path of the target address text.

[0173] Furthermore, step S4 specifically includes:

[0174] After converting the address elements contained in the association path into a computable format, based on the node2vec random walk strategy, random sampling is performed with each node in the association path as the starting point to obtain a node sequence, and the address embedding vector corresponding to the node sequence is generated through the skip-gram model;

[0175] The address embedding vector is used as the input of the preset graph attention network model, and the preset graph attention network model outputs the address feature vector, which integrates the address semantics and address hierarchical structure information.

[0176] Furthermore, step S4 further includes:

[0177] The address feature vector is stored in the vector database.

[0178] Furthermore, step S5 specifically includes:

[0179] Using IVF-FLAT method to search for the nearest address vector of at least one address feature vector in the vector database;

[0180] The matching degree between the neighbor address vector and the address feature vector is calculated based on the preset matching model. The preset matching model is specifically:

[0181]

[0182] Among them, r1 and r2 are preset harmonic coefficients, represents the similarity score between the address feature vector and the neighboring address vector, It represents the overlap score between the address feature vector and its neighboring address vector. G is the function for calculating the number of overlaps between address pair elements. L is the function for calculating the length of address elements corresponding to the address feature vector. k is the type of preset key address element.

[0183] An embodiment of the present application also provides a computer-readable storage medium for storing program code, wherein the program code is used to execute any one of the implementation methods of an address fuzzy matching method described in the aforementioned embodiments.

[0184] In the embodiments of the present application, a method, apparatus, device and readable storage medium for address fuzzy matching are provided. The address elements of the target address text are extracted to construct an address graph. The address feature vector that integrates the address semantics and structural information is extracted using a graph attention network model. This method can better cope with non-standard addresses. At the same time, a vector database is used to perform a matching search of neighboring address vectors without considering time-consuming issues. Finally, the desired fuzzy matching address is screened out according to a preset threshold. This solves the technical problems of existing address matching methods, such as reliance on a standardized address database and manually exhaustive address structures, difficulty in coping with non-standard addresses, and limited matching accuracy. In addition, when faced with large-scale data processing, pair-by-pair calculations are time-consuming and difficult to adapt to all scenarios.

[0185] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0186] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0187] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0188] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0189] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0190] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0191] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store program codes.

[0192] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for fuzzy address matching, characterized in that: include: Get the target address text to be matched; After preprocessing the target address text, the recalled address is screened and matched based on a preset POI database and a preset rule template; If the screening match fails, the target address text is parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing results; Based on a preset graph attention network model and in combination with the address graph, vector encoding and feature extraction are performed on the target address text to obtain an address feature vector of the target address text; Searching for at least one neighboring address vector of the address feature vector in a vector database, and calculating a matching degree between the neighboring address vector and the address feature vector based on a preset matching model; If the matching degree is higher than a preset threshold, the address information corresponding to the neighbor address vector is used as the fuzzy matching result of the target address text matching.

2. The address fuzzy matching method according to claim 1, characterized in that: After the target address text is preprocessed, the recalled address is screened and matched based on a preset POI database and a preset rule template, including: Cleaning the target address text and preprocessing it to a uniform format; Matching the preprocessed target address text with a preset POI database to obtain a number of matching recall addresses; It is determined whether the target address text and the matched recalled addresses have the same POI element or the same address classification information.

3. The address fuzzy matching method according to claim 2, characterized in that: After the target address text is preprocessed, the recalled address is screened and matched based on a preset POI database and a preset rule template, and then the following is further included: If the target address text and the matched recalled addresses have the same POI element or the same address classification information, the matched recalled addresses are element parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing results.

4. The address fuzzy matching method according to claim 2, characterized in that: If the screening match fails, the target address text is parsed based on the entity recognition model, and an address graph is constructed according to the address element parsing result, including: If there is no recall address with the same POI element or the same address classification information among the target address text and the matched plurality of the recall addresses, then based on the Bert model pre-trained in the address entity corpus, the target address text is element parsed to obtain the address element parsing result of the target address text; According to the parsing result of the address element, an address graph is constructed with the address element as a parent node to determine the associated path of the target address text.

5. The address fuzzy matching method according to claim 4, characterized in that: The preset graph attention network model is combined with the address graph to perform vector encoding and feature extraction on the target address text to obtain the address feature vector of the target address text, including: After converting the address elements contained in the association path into a computable format, random sampling is performed based on the node2vec random walk strategy, starting from each node in the association path to obtain a node sequence, and the address embedding vector corresponding to the node sequence is generated through the skip-gram model; The address embedding vector is used as the input of a preset graph attention network model, and the preset graph attention network model outputs an address feature vector, which integrates address semantics and address hierarchical structure information.

6. The address fuzzy matching method according to claim 5, characterized in that: The step of searching a vector database for a neighboring address vector of at least one of the address feature vectors, and calculating a matching degree between the neighboring address vector and the address feature vector based on a preset matching model, specifically includes: Using the IVF-FLAT method, searching for a neighboring address vector of at least one of the address feature vectors in a vector database; The matching degree between the neighbor address vector and the address feature vector is calculated based on a preset matching model.

7. The address fuzzy matching method according to claim 6, characterized in that: The preset matching model is specifically: respectively calculating a similarity score between the address feature vector and the neighboring address vector, and a coincidence score between the address feature vector and the neighboring address vector; The similarity score and the coincidence score are summed based on a preset reconciliation coefficient to obtain a matching degree between the neighbor address vector and the address feature vector.

8. An address fuzzy matching device, characterized in that: include: An acquisition unit, used for acquiring a target address text to be matched; A first matching unit, configured to pre-process the target address text and then perform screening and matching of the recalled addresses based on a preset POI database and a preset rule template; A first processing unit, configured to perform element parsing on the target address text based on an entity recognition model if the screening match fails, and construct an address graph according to the address element parsing result; A feature extraction unit, configured to perform vector encoding and feature extraction on the target address text based on a preset graph attention network model and in combination with the address graph, so as to obtain an address feature vector of the target address text; A second matching unit, configured to search a vector database for a neighboring address vector of at least one of the address feature vectors, and calculate a matching degree between the neighboring address vector and the address feature vector based on a preset matching model; The second processing unit is configured to use the address information corresponding to the neighboring address vector as a fuzzy matching result of the target address text matching if the matching degree is higher than a preset threshold.

9. An address fuzzy matching device, characterized in that: The device comprises a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the address fuzzy matching method according to any one of claims 1 to 7 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program code, and the program code is used to execute the address fuzzy matching method according to any one of claims 1 to 7.

Images