Information processing method and device, computer equipment and storage medium

Abstract

The invention provides an information processing method and device, computer equipment and a storage medium. The method comprises the steps that to-be-processed content is acquired; using an information extraction model to perform first target content identification and second target content identification on the to-be-processed content, the first target content including address content, the second target content including description information, and the description information being information except contact information and the address content; the recognized first target content and the recognized second target content are output, and the first target content and the second target content are displayed in a distinguished mode.

Description

Technical Field

[0001] The present disclosure relates to the technical field of data extraction, and in particular to an information processing method, apparatus, computer equipment and storage medium. Background Art

[0002] The gradual improvement of positioning technology and computer processing technology has greatly facilitated users' daily travel and logistics. In related technologies, users can use navigation APP (Application) to navigate to the target address, or they can edit the delivery address when shopping on the shopping platform to realize online shopping.

[0003] However, in the scenario where users enter information, the original input information needs to be converted into structured information. However, as user needs increase, the content contained in the input information is increasing. Too much information will lead to inaccurate information recognition, seriously affecting the user experience. Summary of the invention

[0004] In view of this, the present disclosure proposes an information processing method, apparatus, computer device and storage medium to solve or partially solve the above problems. This summary of the invention is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0005] Based on the above purpose, the first aspect of the present disclosure provides an information processing method, including:

[0006] Get the content to be processed;

[0007] Using the information extraction model to perform first target content recognition and second target content recognition on the content to be processed, wherein the first target content includes address content, and the second target content includes description information, and the description information is information other than contact information and the address content;

[0008] The identified first target content and second target content are output, wherein the first target content and the second target content are displayed distinguishably.

[0009] In a second aspect of the present disclosure, there is provided an information processing device, comprising:

[0010] The first module is used to obtain the content to be processed;

[0011] A second module is used to use the information extraction model to perform first target content recognition and second target content recognition on the content to be processed, wherein the first target content includes address content, and the second target content includes description information, and the description information is information other than the contact information and the address content;

[0012] The third module is configured to output the identified first target content and the second target content, wherein the first target content and the second target content are displayed separately.

[0013] In a third aspect of the present disclosure, a computer device is provided, 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 method described in the first aspect is implemented.

[0014] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein the non-transitory computer-readable storage medium stores computer instructions, wherein the computer instructions are used to enable a computer to execute the method described in the first aspect.

[0015] As can be seen from the above, the present disclosure provides an information processing method, device, computer equipment and storage medium, which, after obtaining the content to be processed, identifies the first target content and the second target content of the content to be processed, wherein the second target content is the remark information or prompt information other than the contact information and the address content, so as to distinguish the first target content and the second target content in the content to be processed, thereby effectively extracting some remark information, prompt information, etc. filled in by the user, so that the first target content contains only the address information as much as possible, and improves the accuracy of identifying the address information, etc. At the same time, since the second target content is extracted, the user's remark information and other information can be clarified, so that the operator can perceive the user's specific needs faster and more accurately, improve the service quality of the operator, and relatively improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or related technologies, the drawings required for use in the embodiments or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A schematic diagram of an exemplary system according to an embodiment of the present disclosure is shown.

[0018] Figure 2 A schematic diagram is shown of an exemplary input interface for content to be processed and a display interface for identification information output by a model according to an embodiment of the present disclosure.

[0019] Figure 3 A flowchart of an exemplary method according to an embodiment of the present disclosure is shown.

[0020] Figure 4 A schematic diagram of an exemplary apparatus according to an embodiment of the present disclosure is shown.

[0021] Figure 5 A schematic diagram of an exemplary computer device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of this specification more clear, this specification is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0023] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the examples of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Including" or "including" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0024] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0025] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0026] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0027] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0028] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0029] In order to make the purpose, technical solutions and advantages of this specification more clear, this specification is further described in detail below in combination with specific examples and with reference to the accompanying drawings.

[0030] Figure 1 FIG. 1 is a schematic diagram of an exemplary system 100 according to an embodiment of the present disclosure. The system 100 may be a system for implementing an information processing method.

[0031] like Figure 1 As shown, taking the case where a terminal device and a server jointly execute the object display method as an example, the system 100 may include a terminal device 102, a server 104, and a database server 106. The terminal device 102 and the server 104 are connected via a network, such as a wired or wireless network connection. Optionally, the device for implementing the information processing method may be integrated in the terminal device 102. The database server 106 and the server 104 are connected via a network, such as a wired or wireless network connection. The database server 106 may store various data related to the execution of the information processing method, such as address information parameters, generation algorithms, and the like.

[0032] The terminal device 102 may be installed with various applications (Application, APP, Chinese abbreviation: application), such as map service applications, video conferencing applications, life service applications, reading applications, video applications, social applications, payment applications, web browsers and instant messaging tools, etc., all of which can be used for information processing methods and / or displaying processed information. As an optional example, the application (APP) installed on the terminal device 102 may be downloaded and installed from the server 104.

[0033] The terminal device 102 here can be hardware or software. When the terminal device 102 is hardware, it can be various electronic devices with a display screen, including but not limited to smart phones, tablet computers, e-book readers, MP3 players, laptop computers (Laptops) and desktop computers (PCs), etc. When the terminal device 102 is software, it can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (for example, to provide distributed services), or it can be implemented as a single software or software module, which is not specifically limited here.

[0034] The server 104 may be a server that provides various services. For example, it may be a background server that provides support for various applications displayed on the terminal device 102. The database server 106 may also be a database server that provides various services. It is understood that if the server 104 can implement the relevant functions of the database server 106, the database server 106 may not be provided in the system 100.

[0035] Here, the server 104 and the database server 106 can also be hardware or software. When they are hardware, they can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When they are software, they can be implemented as multiple software or software modules (for example, to provide distributed services), or as a single software or software module. No specific limitation is made here.

[0036] It should be noted that the information processing method provided by the example of the present disclosure can be executed by the system 100. Specifically, it can be executed interactively between the terminal device 102, the server 104 and the database server 106. It is understood that in the case where the terminal device 102 has the functions required by the server 104 and the database server 106 to execute the information processing method, the terminal device 102 can also execute it alone. It should be understood that Figure 1 The number of terminal devices 102, servers 104, database servers 106 and users 108 in the figure is only for illustration. According to actual needs, there may be any number of terminal devices, users, servers and database servers.

[0037] In an exemplary application scenario, the user 108 may input the content to be processed through the terminal device 102, and the server 104 may execute the information processing method based on the content to be processed, thereby providing the user 108 with information processing services, and display the results of the information processing services on a page through the terminal device 102. Figure 2As shown, the user can input information into the terminal device 102 in the input area of ​​the content to be processed. After clicking to identify, the terminal device 102 or the server 104 can process the content to be processed and finally display the result on the terminal device 102, which is the consignee, telephone number, location, detailed address, etc. above the input area.

[0038] In some examples, as users' requirements for products or services increase, they will make more personalized requirements based on their own requirements and situations. These requirements are generally not universal and difficult to summarize into unified fields. For example, some users will request not to deliver on weekends, and some will choose the logistics contractor. In some examples, a solution is provided to copy a large section of content to be processed directly into the recognition frame for information recognition. It can extract information from this section of content to be processed to obtain user information, address information, etc. At this time, if the user writes some personalized requirements in the content to be processed, according to the above examples, these requirements may not be directly recognized, but these notes are directly regarded as part of the address information, resulting in the recognition of non-standard information or even recognition errors, which seriously undermines the experience.

[0039] Furthermore, in some examples, such as Figure 3 As shown, another information processing method 300 is exemplarily provided. The method 300 is used for information processing and can be implemented by the terminal device 102 alone or by the system 100. Figure 3 As shown, method 300 may include the following steps.

[0040] Step 302: Obtain content to be processed.

[0041] In this step, the content to be processed may be a piece of content information input by the user through a corresponding input means, or may be content from other devices or programs. Figure 2 As shown, an input area is provided for the user, and the user can write the content to be input into the input area by direct input or copy / paste, etc. After the user clicks the confirmation button such as submit or identify, the content in the input area is regarded as the content to be processed. The user can also be allowed to input by voice or other means, and after receiving a paragraph of user input, the text recognition is performed on it to generate the content to be processed. The user can also be allowed to input by image scanning or other means, and a piece of information is extracted by scanning the object as the content to be processed. In this and subsequent examples, the content to be processed generated by the user inputting text through the input box is used as an example for illustrative description, and no specific limitation is made to this.

[0042] In some examples, with the diversification of user input methods and the personalization of input habits, and even the compatibility issues between different formats of text, the formats of the content to be processed in the input are all-inclusive. In the content to be processed input by the user, there may be typos, spelling mistakes, inconsistent units, incorrect information, etc. And due to different personal Chinese and English input habits, there may be a mix of simplified and traditional Chinese characters, and non-standard capitalization of English words. Even with the personalized settings of the text input method, the content input by the user may carry some special symbols or even pictures, such as Figure 2 the content within the input area shown. All these are not conducive to the information confirmation of the content to be processed, and often there will be situations where it cannot be recognized and the user needs to re-identify it. Therefore, after the user inputs a piece of content to be processed and submits it, it is possible to check whether the input content conforms to the corresponding specifications, such as checking for typos, extra words, etc. If the content to be processed conforms to the specifications and is complete, then it can be directly considered as standard and normalized content, and information extraction can be directly performed; while if there is at least one non-standard place, then it is necessary to further rectify the non-standard place and adjust it to the standard content.

[0043] As described in the foregoing examples, in some scenarios, with the user's personalized settings, the content to be processed may contain more non-standard information. For example, for Chinese information, there may be a mixed arrangement of simplified and traditional Chinese characters; for foreign language information (such as English, French, German, Russian, etc.), there may be a chaotic situation of capitalization; with the addition of more and more personalized symbols, the same word may have more expression forms, such as "two, 2, 貮, ②, etc.", informal languages such as Internet slang; and due to different input habits of different users, the same name may also correspond to multiple expression forms, such as "Building No. 2" may be written as "2#", "2 buildings", "2 blocks", etc. Furthermore, in order to be able to recognize these contents as well, more recognition rules can be added on the basis of the previous recognition methods. It is possible to form a checklist, etc., based on the specific situations in the specific scenarios, and write the specific keywords, words, and symbols to be recognized into it. When the corresponding content is found, it is determined that the content to be processed contains non-standard content. Specifically, such as Figure 2As shown, for the content to be processed in the input area, it at least includes non-standard content such as "Chuangye Road", "2#", "Technology Circle", "② Building", "Building", etc. These non-standard contents need to be further standardized. It can be processed directly after one non-standard content is determined. Of course, in some examples, after all non-standard content in the entire section of content to be processed is determined, each non-standard content is processed. The content to be processed that has completed the standardized processing is the standardized content to be processed. That is, in some examples, after obtaining the content to be processed, the method also includes: performing a standardized check on the content to be processed; in response to the content to be processed including at least one non-standard content, performing standardized processing on the at least one non-standard content to obtain the standardized content to be processed.

[0044] In some examples, the process of normalizing non-standard content can be completed through a corresponding configuration table, which can correspond to the check table in the aforementioned example, or be the same table. When a non-standard content recorded in a table appears in the content to be processed, it can be directly adjusted according to the adjustment method in the table. In other examples, corresponding detection rules can also be designed. When a certain rule is met, corresponding adjustments can be made, such as full-width to half-width, HTML (Hyper Text Markup Language) entity deletion, letter case conversion, simplified and traditional text conversion, etc. That is, in some examples, the normalization of at least one non-standard content includes at least one of the following: deleting the content belonging to the Hypertext Markup Language entity in the at least one non-standard content; converting the foreign content in the at least one non-standard content to uppercase and lowercase; converting the traditional Chinese content in the at least one non-standard content to simplified Chinese; converting the content with a half-width character width in the at least one non-standard content to full-width. Alternatively, in some examples, the normalizing the at least one non-standard content includes: deleting and / or replacing the content in the at least one non-standard content that complies with corresponding rules according to a rule configuration table.

[0045] In a more specific application scenario, the inspection and normalization processing of the content to be processed can specifically include several stages: First, it can be directly adjusted according to rules. For example: removing HTML entities, converting to uppercase, simply removing the leading and trailing characters of a sentence, normalizing the writing of "-", converting full-width characters to half-width characters, converting traditional Chinese to simplified Chinese, etc. After that, corresponding deletion or conversion operations can be performed according to a pre-set inspection list or configuration table. For example, converting the numerical characters before "office, do, support do, building, number, lane, building, shop, phase, room, building, area, building, workshop, block, group, team, primary school, industrial park, building, cross street, rural road, national road, provincial road, row, unit, door, etc." to Arabic numerals; replacing special symbols, characters, and words, such as " replace with 中", "⑤ replace with 5", "『 replace with (", "1# replace with 1号", "| building replace with 1 building", etc.; or deleting useless symbols, words, etc., such as deleting parentheses, deleting "thank you", "contact by phone", etc.; or performing spelling mistake inspection and correction, extra character inspection and deletion, missing character inspection and supplementation, etc. It can be seen that the specific normalization processing methods can be: converting full-width characters to half-width characters for the address, deleting HTML entities, converting letters to uppercase, removing and repairing symbols, regular expression replacement, regular expression deletion, etc., which can be specifically executed according to pre-set rules or comparison tables. Finally, the content to be processed such as Figure 2 shown as "Chuangye Road, Nanshan District, Shenzhen City 2# Technology Park ② Building Building" can be adjusted to "Chuangye Road, Nanshan District, Shenzhen City 2nd Technology Park 2nd Building".

[0046] Step 304, use the information extraction model to perform first target content recognition and second target content recognition on the content to be processed, where the first target content includes address content, and the second target content includes descriptive information, and the descriptive information is information other than contact information and the address content.

[0047] In this step, after obtaining the content to be processed, the information extraction model can be used to extract the required information. Here, the first target content and the second target content will be identified, wherein the first target content can mainly include address content, that is, it can be some common address information. For the content to be processed, in addition to the address information, it may also contain other information, such as contact information, contact phone information, etc., which are generally targeted for identification. However, in some scenarios, more and more users will add some personalized requirements to the content to be processed. These requirements are generally used to describe some specific needs of users, and then they can be called description information. These personalized description information is the second target content. The information extraction model here can extract information by segmenting the standardized content through the retrieval of key fields. Specifically, the information extraction model can be a named entity recognition (NER) model, a relation extraction (RE) model, an event extraction (EE) model, an information extraction model based on a large language model (LLMs), and the like.

[0048] In some more specific examples, the corresponding information extraction model can be used to extract information that is strongly related to the scene, such as the contact object's name, province, city, district, street, detailed address, etc. In other examples, users usually use notes or descriptive information to express their own personalized requirements, and then the descriptive information cannot be extracted through the extraction method of the aforementioned information extraction model. In order to overcome this problem, when training the information extraction model, the corresponding content and corresponding labels can be added, that is, a descriptive information entity is added, and then a large amount of descriptive information text is collected and added to the training set for training, so that the information extraction model can finally extract all the above information when extracting information, for example, it can extract information according to the seven-level entities of name, province, city, district, street, detailed address and descriptive information.

[0049] In some examples, a corresponding training sample set can be set for the situation where there is non-standard information such as remark information in the information to be processed, so as to train the information extraction model so that it can identify the corresponding remark information. In some examples, the training sample set can include a first sample, such as a paragraph of text marked with a corresponding information label, etc. For the first sample, it can include the aforementioned first target content and second target content, and different labels can be associated with the two contents, for example, the first target content is associated with the first label, and the second target content is associated with the second label. In this way, after training with such a training sample set, the information extraction model can extract the first target content and the second target content from the content to be processed respectively. That is, in some examples, the information extraction model is obtained in the following manner: a training sample set is set, the sample set includes a first sample, the first sample includes a first target content and a second target content, the first target content is associated with a first label, and the second target content is associated with a second label; based on the training sample set, the information extraction model is trained so that the information extraction model outputs the first target content and the second target content.

[0050] Further, according to the foregoing, in addition to being able to extract address information and description information, the information extraction model can generally also extract contact information, including the extraction of the contact object name and the contact information (telephone, etc.) of the contact object. In some examples. When extracting name information, it is generally required that the content to be processed input by the user contains the real name of a natural person. However, in the corresponding scenarios, more and more users want to use network names (net names, etc.) as objects of information extraction. In addition, the current information extraction model is not friendly to corporate or legal users. Therefore, similar to the foregoing, when training the information extraction model, the label corresponding to the contact object name (the third label is used to represent the label in the subsequent examples) can be adjusted, and a large number of network names and corporate names can be added to the training set, and marked with the third label. The information extraction model trained in this way can be more flexible in identifying the contact object name. As Figure 2 As shown, the name information "last night's stars" input by the user can be identified and filled into the consignee column. That is, in some examples, the first sample also includes a third target content, and the third target content includes the name of the contact object; the setting of the training sample set includes: expanding the third target content, adding the network name and the company name to the third target content, and associating a third label with the contact object name, the network name and the company name. Among them, the third target content can correspond to the name information of the contact object.

[0051] In some examples, it can be seen from the above content that the information extraction model can be the various models mentioned in the above examples. However, considering the context understanding, sequence modeling, label dependency modeling, efficiency and accuracy, model adaptability, etc., and combined with the needs of the scene, for the scene of address information extraction, the named entity recognition (NER) model can be used for information extraction. Afterwards, for the named entity recognition (NER) model, its specific model structure can be a hole convolution model (Iterated Dilated Convolutional Neural Network, IDCNN) combined with a conditional random field structure (Conditional Random Field, CRF), that is, IDCNN+CRF. IDCNN is an optimization of CNN in sequence labeling tasks. Traditional CNN performs convolution and pooling on adjacent nodes. Hollow convolution means that the nodes selected for convolution may not be adjacent. Traditional CNN will cause the model to ignore context information. The convolution method of IDCNN can make the model field of view wider and fully consider the context semantics. Fast training speed, fast prediction speed, batch calculation, and high quality of training sets can achieve good results.

[0052] Furthermore, in order to improve the model capability of the named entity recognition model and improve the overall recognition accuracy, at least two fully connected layers can be added after the dilated convolution model to increase the model parameters. The model with more model parameters will directly affect the model effect and have a direct impact on the recognition accuracy. In addition, since more model parameters will also lead to lower efficiency of result output, considering the balance between the two, in some examples, two fully connected layers can be added. Of course, the specific number of fully connected layers added can be adjusted according to the needs of the specific scenario. That is, in some examples, the model structure of the information extraction model is a dilated convolution model combined with a conditional random field structure; the information extraction model adds at least two fully connected layers after the dilated convolution model.

[0053] In some examples, for the second target content, since its essence is the description information added by the user to the content to be identified according to personal needs, the second target content may appear at any position of the content to be processed and the number is not fixed. For example, a description information of "do not call on weekends" is added after the phone number, and then a description information of "do not deliver on weekends, choose A logistics for transportation" is added after the address content. In order to further accurately identify the user's intention, it helps the operator to perceive the user's needs more accurately. When at least two second target contents are identified by using the information extraction model, these contents can be directly extracted, and for the convenience of display, these second target contents can be sorted, such as directly connected, or some spacing symbols are added between different second target contents, such as adding a space or adding semicolons, commas, etc. That is, in some examples, the second target content identification of the content to be processed by using the information extraction model includes: in response to identifying at least two second target contents in the content to be processed, extracting and sorting the at least two second target contents.

[0054] Furthermore, for the second target content, different users may be accustomed to adding some symbols to distinguish the second target content from the address content or contact information. For example, the user may add symbols such as brackets and quotation marks to the second target content, and these symbols have no other meaning except for distinguishing the content. However, if they are not processed, they may affect the viewing experience of the operator when viewing the second target content. Furthermore, when arranging the second target content, some cleaning operations of setting symbols may be performed on it, and then the specific content may be spliced. That is, in some examples, the extraction and arrangement of the at least two second target contents includes: symbol cleaning and content splicing processing of the at least two second target contents.

[0055] Step 306: output the identified first target content and second target content, wherein the first target content and the second target content are displayed separately.

[0056] In this step, after the first target content and the second target content are identified, the output operation can be performed on these contents. Generally, the first target content and the second target content can be displayed and output. Since the two parts of the content are distinguished, they can be displayed separately in different areas, that is, the first target content and the second target content can be displayed separately.

[0057] In some examples, in order to facilitate operators to view all the identified information, all this information can be displayed in one display interface, and then the display interface can be divided to form multiple display areas, each type of information corresponds to a display area, and then the first target content can be displayed in the first display area, and the second target content can be displayed in the second display area. Of course, in other examples, the first display area and the second display area can also be presented in the form of two different display windows, or located in different interfaces or web pages, etc. The above is only an exemplary description of the display area, and no specific limitation is made to this. That is, in some examples, the first target content and the second target content are displayed separately, including: displaying the first target content in the first display area; displaying the second target content in the second display area.

[0058] In some examples, for the first target content to be output, it is difficult for the user to complete the content in a very standard format when inputting, and the user may also make input errors, so it is necessary to determine whether the first target content output by the model needs to be completed and / or adjusted. Figure 2 The content to be processed input by the user is shown as missing "province information, street information", etc., and its "district information" is also wrong. As for the first target content that needs to be completed and / or adjusted, the first target content will be searched and identified according to the database when it is identified, such as Figure 2 The "Technology Park No. 2 Chuangye Road" shown. For a database, it generally stores data in a relatively strict hierarchical structure. Specifically, for a database in the field of address recognition, when storing information, it will be located layer by layer according to a standard hierarchical structure to finally determine the storage location of the information. For example, an address will strictly follow the tree structure, first find the "province" node, then determine the "city" node under the "province" node, and so on, and finally store the specific address under the "street" node. In this way, when it is determined that the first target content needs to be completed and / or adjusted, the first target content can be directly completed and adjusted with the storage information of the corresponding data stored in the database (in this scenario, it can mainly be used with the help of its hierarchical information, etc.) to obtain the complete division of the address. That is, in some examples, before the output of the identified first target content, the method also includes: determining whether the first target content needs to be completed and / or adjusted; in response to the need for completion and / or adjustment, searching in the database according to the first target content to determine the storage information of the corresponding content in the database; completing and / or adjusting the first target content according to the storage information.

[0059] In some examples, telephone information may appear in the content to be processed. Since the data structure of telephone information is relatively uniform, regular expressions can be used to extract telephone information. Regular expressions are a pattern used to match character combinations in strings. They can be used to search, edit, and replace text that meets a specific pattern. Regular expressions identify specific information in documents and data by defining the arrangement of characters.

[0060] Finally, the output first target content and second target content can be displayed on the corresponding device (such as the terminal device 102) to give the operator corresponding feedback. It can also be used to store, display, use or reprocess these output contents. According to different application scenarios and implementation needs, the specific output method of these output contents can be flexibly selected.

[0061] For example, for an application scenario in which the method of this example is executed on a single device, these output contents can be directly displayed on the display component (display, projector, etc.) of the current device, so that the operator of the current device can directly see the contents of these output contents from the display component.

[0062] For another example, in an application scenario where the method of this example is executed on a system composed of multiple devices, these output contents can be sent to other preset devices as receivers in the system, that is, synchronization terminals, through any data communication method (wired connection, NFC, Bluetooth, wifi, cellular mobile network, etc.), so that the synchronization terminals can process them later. Optionally, the synchronization terminal can be a preset server, which is generally set up in the cloud as a data processing and storage center, which can store and distribute these output contents; wherein the recipients of the distribution are terminal devices, and the holders or operators of these terminal devices can be testers of map applications, relevant personnel of online stores, supervisors of network platforms, etc.

[0063] For another example, in an application scenario where the method of this example is executed on a system consisting of multiple devices, the output contents can be sent directly to a preset terminal device through any data communication method, and the terminal device can be one or more of the ones listed in the preceding paragraphs.

[0064] From the above embodiments, it can be seen that the information processing method provided by the embodiments of the present disclosure, after obtaining the content to be processed, identifies the first target content and the second target content of the content to be processed, wherein the second target content is the remark information or prompt information other than the contact information and the address content, so as to directly distinguish the first target content and the second target content in the content to be processed, thereby effectively extracting some remark information and prompt information filled in by the user, so that the first target content contains only the address information as much as possible, and improves the accuracy of identifying the address information, etc. At the same time, since the second target content is extracted, the user's remark information and other information can be clarified, so that the operator can perceive the user's specific needs faster and more accurately, improve the service quality of the operator, and relatively improve the user experience.

[0065] In specific application scenarios, according to the exemplary method 300 of the present application, more accurate recognition can be achieved, the speed at which users fill in addresses can be accelerated, the time spent on filling in addresses can be reduced, users can fill in addresses repeatedly, and the user experience can be improved. At the same time, the description information is identified, and a cleaner and more accurate detailed address can be returned. For the merchant side, there are some scenarios that rely on the description information filled in by the user. For example, if the description information involves a designated logistics company, the merchant can specify the logistics company to ship the goods when shipping. This function can enhance the user experience of filling in addresses, and allow the merchant side to perceive faster and more accurately, avoiding the loss of user experience caused by inadequate perception of description information during merchant shipment, user returns, and logistics distribution, and reducing customer complaints.

[0066] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the embodiment of the present disclosure can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0067] It should be noted that the above is a description of a specific embodiment of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0068] Based on the same technical concept, corresponding to any of the above-mentioned embodiment methods, the present disclosure also provides an information processing device.

[0069] refer to Figure 4 , the information processing device comprises:

[0070] The first module 410 is used to obtain content to be processed.

[0071] The second module 420 is used to use the information extraction model to perform first target content identification and second target content identification on the content to be processed, wherein the first target content includes address content, and the second target content includes description information, and the description information is information other than contact information and the address content.

[0072] The third module 430 is configured to output the identified first target content and second target content, wherein the first target content and the second target content are displayed separately.

[0073] In some exemplary embodiments, the third module 430 is further configured to:

[0074] Displaying the first target content in a first display area;

[0075] The second target content is displayed in the second display area.

[0076] In some exemplary embodiments, the information extraction model is obtained by:

[0077] Setting a training sample set, the sample set including a first sample, the first sample including a first target content and a second target content, the first target content being associated with a first label, and the second target content being associated with a second label;

[0078] The information extraction model is trained based on the training sample set, so that the information extraction model outputs the first target content and the second target content.

[0079] In some exemplary embodiments, the first sample further includes a third target content, wherein the third target content includes a contact object name;

[0080] The step of setting a training sample set includes:

[0081] The third target content is expanded, a network name and a company name are added to the third target content, and a third tag is associated with the contact object name, the network name and the company name.

[0082] In some exemplary embodiments, the model structure of the information extraction model is a dilated convolutional model combined with a conditional random field structure; the information extraction model adds at least two fully connected layers after the dilated convolutional model.

[0083] In some exemplary embodiments, the first module 410 is further configured to:

[0084] Performing a standardized check on the content to be processed;

[0085] In response to the content to be processed including at least one non-standard content, the at least one non-standard content is standardized to obtain standardized content to be processed.

[0086] In some exemplary embodiments, the first module 410 is further used for at least one of the following:

[0087] Deleting the content belonging to the hypertext markup language entity in the at least one non-standard content;

[0088] Converting the foreign language content in the at least one non-standard content into uppercase and lowercase letters;

[0089] Converting the Traditional Chinese content in the at least one non-standard content into Simplified Chinese;

[0090] The half-width characters in the at least one non-standard content are converted to full-width.

[0091] In some exemplary embodiments, the third module 430 is further configured to:

[0092] Determining whether the first target content needs to be completed and / or adjusted;

[0093] In response to the need for completion and / or adjustment, searching in a database according to the first target content to determine storage information of the corresponding content in the database;

[0094] The first target content is completed and / or adjusted according to the stored information.

[0095] In some exemplary embodiments, the second module 420 is further configured to:

[0096] In response to identifying at least two second target contents in the content to be processed, the at least two second target contents are extracted and sorted.

[0097] In some exemplary embodiments, the second module 420 is further configured to:

[0098] The at least two second target contents are subjected to symbol cleaning and content splicing processing.

[0099] For the convenience of description, the above devices are described in terms of functions divided into various modules. Of course, when implementing the embodiments of the present disclosure, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0100] The device of the above embodiment is used to implement the corresponding method 300 in the above embodiment, and has the beneficial effects of the corresponding method 300 embodiment, which will not be described in detail here.

[0101] Based on the same technical concept, corresponding to any of the above-mentioned embodiments and methods, the embodiment of the present disclosure further provides a computer device for implementing the above-mentioned method 300. Figure 5 FIG. 5 is a schematic diagram showing the hardware structure of an exemplary computer device 500 provided in an embodiment of the present disclosure. The computer device 500 can be used to implement Figure 1 In some scenarios, the computer device 500 can also be used to implement Figure 1 Server 104 and database server 106.

[0102] like Figure 5 As shown, the computer device 500 may include: a processor 502, a memory 504, a network module 506, a peripheral interface 508 and a bus 510. The processor 502, the memory 504, the network module 506 and the peripheral interface 508 are connected to each other in communication within the computer device 500 through the bus 510.

[0103] Processor 502 may be a central processing unit (CPU), an image processor, a neural network processor (NPU), a microcontroller (MCU), a programmable logic device, a digital signal processor (DSP), an application specific integrated circuit (ASIC), or one or more integrated circuits. Processor 502 may be used to perform functions related to the technology described in this disclosure. In some embodiments, processor 502 may also include multiple processors integrated into a single logical component. For example, Figure 5 As shown, processor 502 may include a plurality of processors 502a, 502b, and 502c.

[0104] The memory 504 may be configured to store data (eg, instructions, computer code, etc.). Figure 5As shown, the data stored in the memory 504 may include program instructions (e.g., program instructions for implementing the method 300 of the embodiment of the present disclosure) and data to be processed (e.g., the memory may store configuration files of other modules, etc.). The processor 502 may also access the program instructions and data stored in the memory 504, and execute the program instructions to operate on the data to be processed. The memory 504 may include a volatile storage device or a non-volatile storage device. In some embodiments, the memory 504 may include a random access memory (RAM), a read-only memory (ROM), an optical disk, a magnetic disk, a hard disk, a solid-state drive (SSD), a flash memory, a memory stick, etc.

[0105] The network interface 506 can be configured to provide the computer device 500 with communication with other external devices via a network. The network can be any wired or wireless network capable of transmitting and receiving data. For example, the network can be a wired network, a local wireless network (e.g., Bluetooth, WiFi, near field communication (NFC)), a cellular network, the Internet, or a combination thereof. It is understood that the type of network is not limited to the above specific examples.

[0106] The peripheral interface 508 can be configured to connect the computer device 500 to one or more peripheral devices to achieve information input and output. For example, the peripheral devices can include input devices such as a keyboard, a mouse, a touch pad, a touch screen, a microphone, and various sensors, and output devices such as a display, a speaker, a vibrator, and an indicator light.

[0107] The bus 510 may be configured to transmit information between various components of the computer device 500 (e.g., the processor 502, the memory 504, the network interface 506, and the peripheral interface 508), such as an internal bus (e.g., a processor-memory bus), an external bus (USB port, PCI-E bus), etc.

[0108] It should be noted that, although the architecture of the above-mentioned computer device 500 only shows the processor 502, the memory 504, the network interface 506, the peripheral interface 508 and the bus 510, in the specific implementation process, the architecture of the computer device 500 may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the architecture of the above-mentioned computer device 500 may also only include the components necessary for implementing the embodiments of the present disclosure, and does not necessarily include all the components shown in the figure.

[0109] The computer device 500 of the above embodiment is used to implement the method 300 described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be described in detail here.

[0110] Based on the same technical concept, corresponding to any of the above-mentioned embodiments, the present disclosure also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute method 300 described in any of the above embodiments.

[0111] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0112] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the method 300 described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0113] Based on the same technical concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure further provides a computer program product, including computer program instructions, which, when executed on a computer, enable the computer to execute the method 300 described in any of the above-mentioned embodiments. In some embodiments, the computer program instructions may be executed by one or more processors of a computer so that the computer and / or the processor execute the method 300. Corresponding to the execution subject corresponding to each step in each embodiment of the method 300, the processor that executes the corresponding step may belong to the corresponding execution subject.

[0114] The computer program product of the above embodiment is used to enable the computer and / or the processor to execute the method 300 described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be described in detail here.

[0115] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples. Based on the concept of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of simplicity.

[0116] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present disclosure difficult to understand, the known power / ground connections to the integrated circuit (IC) chips and other components may or may not be shown in the provided figures. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present disclosure difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present disclosure will be implemented (that is, these details should be fully within the scope of understanding of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present disclosure, it is apparent to those skilled in the art that the embodiments of the present disclosure can be implemented without these specific details or with changes in these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0117] Although the present disclosure has been described in conjunction with specific embodiments of the present disclosure, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0118] The embodiments of the present disclosure are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. An information processing method, comprising: Get the content to be processed; Using the information extraction model to perform first target content recognition and second target content recognition on the content to be processed, wherein the first target content includes address content, and the second target content includes description information, and the description information is information other than contact information and the address content; The identified first target content and second target content are output, wherein the first target content and the second target content are displayed distinguishably.

2. The method according to claim 1, wherein: The distinguishing and displaying the first target content and the second target content includes: Displaying the first target content in a first display area; The second target content is displayed in the second display area.

3. The method according to claim 1, wherein: The information extraction model is obtained in the following way: Setting a training sample set, the sample set including a first sample, the first sample including a first target content and a second target content, the first target content being associated with a first label, and the second target content being associated with a second label; The information extraction model is trained based on the training sample set, so that the information extraction model outputs the first target content and the second target content.

4. The method according to claim 3, wherein: The first sample further includes a third target content, wherein the third target content includes a contact object name; The step of setting a training sample set includes: The third target content is expanded, a network name and a company name are added to the third target content, and a third tag is associated with the contact object name, the network name and the company name.

5. The method according to claim 1, wherein: The model structure of the information extraction model is a dilated convolutional model combined with a conditional random field structure; the information extraction model adds at least two fully connected layers after the dilated convolutional model.

6. The method according to claim 1, wherein: After obtaining the content to be processed, the method further includes: Performing a standardized check on the content to be processed; In response to the content to be processed including at least one non-standard content, the at least one non-standard content is standardized to obtain standardized content to be processed.

7. The method according to claim 6, wherein: The step of normalizing the at least one non-standard content includes at least one of the following: Deleting the content belonging to the hypertext markup language entity in the at least one non-standard content; Converting the foreign language content in the at least one non-standard content into uppercase and lowercase letters; Converting the Traditional Chinese content in the at least one non-standard content into Simplified Chinese; The half-width characters in the at least one non-standard content are converted to full-width.

8. The method according to claim 1, wherein: Before outputting the identified first target content, the method further includes: Determining whether the first target content needs to be completed and / or adjusted; In response to the need for completion and / or adjustment, searching in a database according to the first target content to determine storage information of the corresponding content in the database; The first target content is completed and / or adjusted according to the stored information.

9. The method according to claim 1, wherein: The using the information extraction model to identify the second target content of the content to be processed includes: In response to identifying at least two second target contents in the content to be processed, the at least two second target contents are extracted and sorted.

10. The method according to claim 9, wherein: The extracting and arranging the at least two second target contents includes: The at least two second target contents are subjected to symbol cleaning and content splicing processing.

11. An information processing device, comprising: The first module is used to obtain the content to be processed; A second module is used to use the information extraction model to perform first target content recognition and second target content recognition on the content to be processed, wherein the first target content includes address content, and the second target content includes description information, and the description information is information other than the contact information and the address content; The third module is configured to output the identified first target content and the second target content, wherein the first target content and the second target content are displayed separately.

12. A computer 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 method according to any one of claims 1 to 10 is implemented.

13. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to make a computer execute the method according to any one of claims 1 to 10.

Images