Time information determination method and apparatus, storage medium, electronic device, and product
By combining search engines and intelligent information extraction technology with generative models, the business hours of POIs are automatically generated, solving the problem of missing or inaccurate business hours of POIs in electronic maps, and achieving efficient and low-cost updates and improved accuracy of business hours.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING BAIDU NETCOM SCI & TECH CO LTD
- Filing Date
- 2022-07-28
- Publication Date
- 2026-06-19
AI Technical Summary
The missing or inaccurate opening hours of POIs in electronic maps affect users' travel decisions and increase decision-making costs, leading to a decline in user experience and user churn.
By combining search engines and intelligent information extraction technology with business hours generation models and incremental learning models, business hours of POIs are automatically generated from Internet intelligence fragments. The accuracy of the generation model is improved through a large-scale concurrent intelligent interactive voice verification system.
It enables automated generation and timely updates of POI business hours, reduces update costs, improves the accuracy and timeliness of the generated model, and reduces resource waste.
Smart Images
Figure CN115238183B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of computer technology, and more particularly to the field of information processing technology, specifically to a method, apparatus, storage medium, electronic device, and product for determining time information. Background Technology
[0002] With the rise of electronic maps, users are increasingly relying on them for travel decisions. Therefore, the construction of Points of Interest (POIs) on electronic maps is extremely important.
[0003] The operating hours of POIs may influence users' travel decisions and their experience with POIs. Summary of the Invention
[0004] This disclosure provides a method, apparatus, storage medium, electronic device, and product for determining time information.
[0005] According to a first aspect of this disclosure, a method for determining time information is provided, the method comprising:
[0006] Obtain time information of the target point of interest (POI) from web search results; generate candidate business hours for the target POI based on the time information; and, in response to confirming that the candidate business hours are verified to be correct, determine the candidate business hours as the business hours of the target POI.
[0007] According to a second aspect of this disclosure, a time information determining apparatus is provided, the apparatus comprising:
[0008] The module is configured to acquire time information of a target point of interest (POI) from web search results; the module is configured to generate candidate business hours for the target POI based on the time information; and the module is configured to determine the candidate business hours as the business hours of the target POI in response to the determination that the candidate business hours are verified as correct.
[0009] According to a third aspect of this disclosure, an electronic device is provided, comprising:
[0010] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of the first aspect.
[0011] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are configured to cause the computer to perform the method according to the first aspect.
[0012] According to a fifth aspect of this disclosure, a computer product is provided, including a computer program that, when executed by a processor, implements the method according to the first aspect.
[0013] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description
[0014] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:
[0015] Figure 1 A schematic diagram of the application environment of this disclosure embodiment is shown;
[0016] Figure 2 A flowchart illustrating a time information determination method provided in an embodiment of this disclosure is shown.
[0017] Figure 3 A schematic diagram showing the location of the business hours attribute of a POI provided in an embodiment of this disclosure is shown;
[0018] Figure 4 A flowchart illustrating a method for obtaining time information according to an embodiment of this disclosure is shown;
[0019] Figure 5 This illustration shows a schematic diagram of obtaining a first quantity of time information according to an embodiment of the present disclosure;
[0020] Figure 6 A flowchart illustrating a method for generating business hours provided in an embodiment of this disclosure is shown;
[0021] Figure 7 A flowchart illustrating a method for structuring time information according to an embodiment of this disclosure is shown;
[0022] Figure 8 A schematic diagram of the information acquisition process according to an embodiment of this disclosure is shown;
[0023] Figure 9 A schematic diagram of a time information determination device provided in an embodiment of this disclosure is shown.
[0024] Figure 10 A schematic block diagram of an example electronic device 1000 that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation
[0025] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.
[0026] With the rise of electronic maps, users are increasingly relying on them for travel decisions. The construction of Points of Interest (POIs) on electronic maps is extremely important, especially the operating hours of those POIs.
[0027] In an electronic map, a Point of Interest (POI) can be understood as a single location. A POI is primarily composed of its own attributes, human perception / understanding of the POI, and potential connections (transactions / business dealings / service provision) between people and the POI. The POI's own attributes mainly include coordinates, name, telephone number, business hours, parent and child points, and tags.
[0028] If POI (Point of Interest) opening hours are missing, it will directly affect users' travel decisions. Users might try to find opening hours by phone or online, but this increases the decision-making cost before traveling. When the accuracy of POI opening hours is low, it will directly impact the user experience and may even lead to user churn.
[0029] Based on this, the present disclosure provides an information acquisition method and apparatus, which automatically generates the business hours corresponding to the POI by acquiring the time information of POI in the webpage, and updates it in real time on a large scale to ensure the timeliness of the business hours, while keeping the update cost low.
[0030] The information acquisition method provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 101 communicates with server 102 via a network. Terminal 101 can obtain the business hours from server 102 and display the business hours in the POI's attribute interface. Terminal 101 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. Server 102 can be a standalone server or a server cluster consisting of multiple servers.
[0031] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.
[0032] This disclosure utilizes search engines and intelligent information extraction technology, combined with a business hours generation model and an incremental learning business hours decision model, to automatically generate business hours for Points of Interest (POIs) from intelligence fragments on the internet. A large-scale concurrent intelligent interactive voice verification system is used as the verification method for the generation model, improving the accuracy of the generated business hours. The following embodiments will describe the information acquisition method and apparatus of this disclosure in conjunction with the accompanying drawings.
[0033] Figure 2 A flowchart illustrating a time information determination method provided in an embodiment of this disclosure is shown, as follows: Figure 2 As shown, the method may include:
[0034] In step S210, the time information of the target POI is obtained from the web search results.
[0035] In this embodiment of the disclosure, by calling the retrieval interface, a search is performed on the webpage based on the keyword combination of the target POI to obtain search results for the target POI.
[0036] From the search results, retrieve time information related to the business hours of the target POI. This time information can be one or more items.
[0037] In step S220, candidate business hours for the target POI are generated based on time information.
[0038] In this embodiment of the disclosure, a pre-set structure for business hours is used to perform structured processing on the acquired time information. In other words, this disclosure can standardize the format of the acquired time information.
[0039] The time information, after being standardized, is further matrixed and then input into a pre-built business hours generation model to determine the candidate business hours of the target POI.
[0040] In this disclosure, the operating hours of the target POI can also be referred to as the POI's operating hours attribute, and its location on the map can be referenced. Figure 3 . Figure 3 A schematic diagram showing the location of the business hours attribute of a POI provided in an embodiment of this disclosure is shown.
[0041] For example, by searching for store A on the map and determining its location, the store A's POI attribute can display its business hours, such as 10:00-22:00 as shown in the figure.
[0042] In step S230, in response to determining that the candidate business hours verification is correct, the candidate business hours are determined as the business hours of the target POI.
[0043] In this embodiment, the generated candidate business hours can be verified using a smartphone or a business hour discrimination model. If the generated candidate business hour is correct, it is acquired and displayed as an attribute of the target POI at the corresponding location. This provides users with decision-making information for their travel.
[0044] The information acquisition method provided in this disclosure allows for the low-cost acquisition of POI business hours information through web search. By structuring the acquired time, the business hours of the target POI can be automatically generated, resulting in highly timely business hours that can be processed concurrently, reducing resource waste.
[0045] The following embodiments of this disclosure will illustrate the time information for obtaining a Point of Interest (POI) from web search results.
[0046] Figure 4 A flowchart illustrating a method for obtaining time information according to an embodiment of this disclosure is shown, as follows: Figure 4 As shown, the method may include:
[0047] In step S410, a combination of search keywords for the POI is constructed.
[0048] In this embodiment, it is necessary to search for the required information on web pages using a search engine. A search engine can be understood as a retrieval technology that, based on user needs and certain algorithms, uses specific strategies to retrieve specified information from the Internet and return it to the user. Search engines rely on various technologies, such as web crawling, search ranking, web page processing, big data processing, and natural language processing, to provide users with fast and highly relevant information services. The core modules of search engine technology generally include crawling, indexing, retrieval, and ranking, and a series of other auxiliary modules can be added to create a better online environment for users.
[0049] In this disclosure, search formats can be constructed based on search queries. For example, a search keyword combination could be a POI name + city + business hours format. Alternatively, a search keyword combination could be a POI name + city + official website format. For example, for Store A, search for "XX Shopping Center Store XX Region Official Website".
[0050] In step S420, based on the combination of search keywords, the target POI is searched in the web search results to obtain the time information of multiple target POIs.
[0051] In this embodiment of the disclosure, the search results for a webpage may include the id of the target POI, the webpage creation timestamp, the webpage creation time, the webpage link, the webpage title, the webpage summary, and the target search keywords.
[0052] In this disclosure, the time information of multiple target POIs can be obtained from all search results.
[0053] In step S430, time information including business hours is obtained from the time information of multiple target POIs.
[0054] In this embodiment of the disclosure, time information containing business hours is determined from multiple searched time information, and a first quantity of time information is extracted. The first quantity can be the first N webpage text results of the extracted target POI, or it can be all the search results on the first page.
[0055] For example, Figure 5 This illustration shows a schematic diagram of obtaining a first quantity of time information according to an embodiment of the present disclosure, such as... Figure 5 As shown, searching for the BC store in the online store yields the first three results regarding its operating hours. The first result states: "Business hours: Monday to Friday: 10:00 AM - 10:00 PM; Saturday and Sunday: 9:30 AM - 10:00 PM." The second result states: "Business hours: Monday to Friday: 10:00 AM - 10:00 PM; Saturday and Sunday: 9:00 AM - 10:00 PM." The third result states: "Business hours: 10:00 AM to 10:00 PM."
[0056] In this embodiment of the disclosure, the target POI includes a first type of POI and a second type of POI. The first type of POI contains telephone information but does not contain business hours; the second type of POI does not contain telephone information but contains business hours. The process of obtaining the business hours and telephone field of the target POI also includes fields such as POI name, POI tag, and POI city.
[0057] In this disclosure, after obtaining the time information of the target POI, the multiple time information entries are expressed in different ways, requiring unification of these different expressions. That is, the multiple time information entries need to be structured. The following embodiments will illustrate how to perform structured processing of time information to obtain the business hours of the target POI.
[0058] Figure 6 A flowchart illustrating a method for generating business hours according to an embodiment of this disclosure is shown, such as... Figure 6 As shown, the method may include:
[0059] In step S610, the first business time segment is obtained from the time information.
[0060] In this embodiment of the disclosure, a second number of first business hours segments can be obtained, where the second number can be the top M time information items from the first number of search results. For ease of distinction, this disclosure refers to the time information segments obtained from webpage text as first business hours segments. As in the above embodiment, the first business hours segment can be: Monday to Friday: 10:00 AM - 10:00 PM; Saturday and Sunday: 9:30 AM - 10:00 PM.
[0061] Understandably, the time information provided for a term refers to a segment of the first business hours. For example, Figure 5 There are three segments of the first business hours.
[0062] In step S620, the first business time segment is processed into a structured form to obtain the second business time segment.
[0063] In this embodiment of the disclosure, for each first business time segment, it is necessary to perform structured processing on the first business time segment based on a predetermined structure. Based on the structured business time segment, a second business time segment is obtained.
[0064] In step S630, the second business hours segment is input into the pre-built business hours generation model to generate candidate business hours for the target POI.
[0065] In this embodiment of the disclosure, the second number of second business time segments obtained are processed and then input into a pre-built business time generation model to obtain the candidate business time of the target POI.
[0066] The business hours generation model is built on the Transformer architecture.
[0067] The training data for the business hours generation model includes samples from the POI database that contain business hours and samples from the POI database that do not contain business hours.
[0068] For a sample *It* containing business hours in the POI database, the corresponding internet business hour segments (i1, ..., iM) can be obtained by searching a search engine. These business hour segments are then input into a generative model to generate the corresponding business hours. The corresponding loss function can be Loss1 = MSE(It, Ig), where *It* represents the sample and *Ig* represents the business hours.
[0069] For samples in the POI database that do not contain operating hours, the corresponding internet operating time segments (i1,...,iM) are obtained by searching a search engine. These operating time segments are then input into a generative model to generate the corresponding operating hours. This data is then input into a discriminative model for judgment, with the expectation that the model will output true. The corresponding loss function is: Loss2 = min1 / 2(D(Ig)–1)^2.
[0070] This disclosure uses the above method to train and learn the business hours generation model, which enables the generation model to generate business hours with higher accuracy.
[0071] In this disclosure, the second business time segment input to the business time generation model is a matrix structured from the first business time segment. The following embodiments will illustrate the structured business time segment.
[0072] Figure 7 A flowchart illustrating a method for structuring time information according to an embodiment of this disclosure is shown, such as... Figure 7 As shown, the method may include:
[0073] In step S710, the operating time period of the target POI is determined and the operating time period is divided into days.
[0074] The time period includes the start time and the end time.
[0075] In step S720, each day is divided into different time periods.
[0076] The time period includes the start time and the end time.
[0077] In step S730, based on the start and end times of different time periods each day, the first business time segment is structurally processed to construct a matrix, thereby obtaining the second business time segment.
[0078] In this disclosure, the structuring of time information can be based on a weekday cycle, with the cycle divided into days, and each day further divided into different time segments. For example, a week can be divided into seven days, and each day into morning and afternoon, resulting in 14 time segments. Each time segment consists of a corresponding start and end time tuple. Figure 5 The first time information in the table can be structured into business hour segments as shown in Table 1. For example, in Table 1, Monday morning is from 10:00 to 12:00 and afternoon is from 12:00 to 22:00.
[0079] Table 1
[0080]
[0081] After structuring the first business time segment, it is also possible to map the first business time segment to the time period to obtain the mapping relationship between the first business time segment and the time period.
[0082] It should be noted that if there are times when the business is not open, the corresponding position is (0, 0).
[0083] Based on the left-to-right order, the obtained mapping relationship is used to construct a matrix representation of time segments, resulting in the second business time segment. For example, in this disclosure, it is a 7x4 time segment matrix. A 7x4 time segment matrix can be understood as 7 days, with two time periods per day, including a start time and an end time, for a total of 4 time points.
[0084] For example, based on Table 1 above, the obtained time segment matrix is as follows:
[0085] [[10,12,12,22],
[0086] [10,12,12,22],
[0087] [10,12,12,22],
[0088] [10,12,12,22],
[0089] [10,12,12,22],
[0090] [9.5,12,12,22],
[0091] [9.5 12,12,22], ]
[0093] It is understandable that this time segment matrix corresponds to a first business time segment.
[0094] In this embodiment of the disclosure, a second number of second business time segments are input into a determined business time generation model to obtain the business time of the target POI. For example, if the second number is M, then M 7x4 time matrices are input into the business time generation model.
[0095] In this embodiment of the disclosure, an operating time discrimination model can also be determined, which is used to verify the accuracy of the operating time generated by the operating time generation model.
[0096] In this disclosure, negative sample training data and their corresponding first labels can be obtained, as well as positive sample training data and their corresponding second labels. Based on the negative sample training data and the first labels, the positive sample training data and the second labels, and a loss function, a pre-built network model is trained to determine the business hours discrimination model. The loss function can be:
[0097] Loss=min1 / 2(D(Iinp)–label)^2
[0098] Where Iinp is the input business hours matrix, and label is the label corresponding to the input business hours matrix.
[0099] The first label can be 0, indicating an incorrect business hours, and the second label can be 1, indicating a correct business hours.
[0100] In some embodiments of this disclosure, the business hours generation model is trained using first business hours sample data, second business hours sample data, and a corresponding loss function. The first business hours sample data consists of POIs containing business hours. The second business hours sample data consists of POIs not containing business hours and business hours verified correctly by the business hours discrimination model.
[0101] In some embodiments of this disclosure, the business hours discrimination model is trained based on negative sample training data and its corresponding first label and loss function, and positive sample training data and its corresponding second label and loss function. The negative sample training data consists of sample POIs with existing business hours, where the existing business hours are inconsistent with the generated business hours, or the negative sample training data consists of sample POIs that do not contain business hours. The positive sample training data consists of sample POIs with existing business hours, where the existing business hours are consistent with the generated business hours.
[0102] In some embodiments of this disclosure, candidate business hours can be identified based on a pre-determined business hour discrimination model to determine a label for the candidate business hours. In response to determining the label as a second label, the candidate business hours are identified as the business hours of the target POI, and the target is obtained.
[0103] In response to determining the label as the first label, the business hours and the first label are added to the negative sample training data of the business hours discrimination model.
[0104] In some embodiments of this disclosure, the obtained business hours can also be determined based on a smartphone call pattern. This utilizes a phone call pattern to perform large-scale, high-concurrency, automated verification of the business hours generated by the generator in the previous step.
[0105] For example, using a smart phone system, business hours are inquired via telephone, and the results are obtained. Upon confirming that the time in the inquiry result matches the business hours, the business hours of the target POI are retrieved.
[0106] In this disclosure, the label corresponding to the business hours of the target POI can also be determined. Furthermore, the business hours and corresponding labels are added to the positive sample training data of the business hours discrimination model.
[0107] By using the intelligent telephone model, the performance of the generative model can be evaluated, compensating for the deficiencies of the business time discrimination model and improving its decision-making ability, thereby indirectly enhancing the generator's generation capability. Furthermore, verified samples automatically flow to the POI database, forming a continuously accumulating sample pool, which greatly increases the training sample size of the generative adversarial model, further improving its ability to generate business times.
[0108] The information acquisition method provided in this disclosure can be referred to... Figure 8 A schematic diagram. Figure 8 A schematic diagram of the information acquisition process according to an embodiment of this disclosure is shown, such as... Figure 8 As shown, the system comprises eight modules: telephone filtering, search format construction, web page text extraction, business hour segment extraction, business hour segment formulation (i.e., structured time information), business hour generation model, business hour discrimination model, and interactive voice verification system. This process enables automated generation and timely updates of business hours, supports large-scale concurrency, allows for historical risk mining, and offers low update costs.
[0109] Based on and Figure 2 The method shown follows the same principle. Figure 9 A schematic diagram of a time information determination device provided in an embodiment of this disclosure is shown, as follows: Figure 9 As shown, the time information determining device 900 may include:
[0110] The acquisition module 901 is used to acquire the time information of the target point of interest (POI) from the web search results; the generation module 902 is used to generate candidate business hours of the target POI based on the time information; and the determination module 903 is used to determine the candidate business hours as the business hours of the target POI in response to the determination that the candidate business hours are verified to be correct.
[0111] In this embodiment of the disclosure, the acquisition module 901 is used to construct a combination of search keywords for POIs; based on the combination of search keywords, search for target POIs in web page search results to obtain time information of multiple target POIs; and from the time information of multiple target POIs, acquire the time information that includes business hours.
[0112] In this embodiment of the disclosure, the generation module 902 is used to obtain a first business time segment from the time information; perform structured processing on the first business time segment to obtain a second business time segment; and input the second business time segment into a pre-built business time generation model to generate candidate business times for the target POI.
[0113] In this embodiment of the disclosure, the generation module 902 is used to determine the operating time period of the target POI and to divide the operating time period into days; for each day, the day is divided into different time periods, the time periods including start time and end time; based on the start time and end time of different time periods each day, the first operating time segment is structurally processed to construct a matrix to obtain the second operating time segment.
[0114] In this embodiment of the disclosure, the business hours generation model is trained using first business hours sample data, second business hours sample data, and a corresponding loss function; the first business hours sample data consists of POIs containing business hours; the second business hours sample data consists of POIs not containing business hours and business hours verified correctly by the business hours discrimination model.
[0115] In this embodiment of the disclosure, the business hours discrimination model is trained based on negative sample training data and its corresponding first label and loss function, and positive sample training data and its corresponding second label and loss function; the negative sample training data are sample POIs with existing business hours, and the existing business hours are inconsistent with the generated business hours, or the negative sample training data are sample POIs that do not contain business hours; the positive sample training data are sample POIs with existing business hours, and the existing business hours are consistent with the generated business hours.
[0116] In this embodiment of the disclosure, the determining module 903 is used to determine the candidate business hours based on a pre-determined business hours discrimination model, and determine the label of the candidate business hours; in response to determining the label as a second label, the candidate business hours are determined as the business hours of the target POI.
[0117] In this embodiment of the disclosure, the determining module 903 is further configured to, in response to determining that the label is the first label, add the business hours and the first label to the negative sample training data of the business hours discrimination model.
[0118] In this embodiment of the disclosure, the determining module 903 is used to make a telephone inquiry about the candidate business hours based on the mode of a smart phone, and obtain an inquiry result; in response to determining that the business hours in the inquiry result are consistent with the candidate business hours, the candidate business hours are determined as the business hours of the target POI.
[0119] In this embodiment of the disclosure, the determining module 903 is further configured to determine the second label corresponding to the business hours of the target POI; and add the business hours and the corresponding label to the positive sample training data of the business hours discrimination model.
[0120] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0121] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.
[0122] Figure 10 A schematic block diagram of an example electronic device 1000 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.
[0123] like Figure 2As shown, device 200 includes a computing unit 201, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 202 or a computer program loaded from storage unit 208 into random access memory (RAM) 203. RAM 203 may also store various programs and data required for the operation of device 200. The computing unit 201, ROM 202, and RAM 203 are interconnected via bus 204. Input / output (I / O) interface 205 is also connected to bus 204.
[0124] Multiple components in device 200 are connected to I / O interface 205, including: input unit 206, such as keyboard, mouse, etc.; output unit 207, such as various types of monitors, speakers, etc.; storage unit 208, such as disk, optical disk, etc.; and communication unit 209, such as network card, modem, wireless transceiver, etc. Communication unit 209 allows device 200 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0125] The computing unit 201 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 201 performs the various methods and processes described above, such as information acquisition methods. For example, in some embodiments, the information acquisition method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 208. In some embodiments, part or all of the computer program may be loaded and / or installed on device 200 via ROM 202 and / or communication unit 209. When the computer program is loaded into RAM 203 and executed by the computing unit 201, one or more steps of the information acquisition method described above may be performed. Alternatively, in other embodiments, the computing unit 201 may be configured to perform the information acquisition method by any other suitable means (e.g., by means of firmware).
[0126] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.
[0127] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0128] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0129] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).
[0130] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.
[0131] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.
[0132] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.
[0133] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.
Claims
1. A method for determining time information, the method comprising: Obtain the time information of the target Point of Interest (POI) from web search results; From the time information, a first business time segment is obtained, which is a time information segment obtained from the web page text; Determine the operating time period of the target POI and divide the operating time period into days; For each day, the day is divided into different time periods, which include start and end times; Based on the start and end times of different time periods each day, the first business time segment is structurally processed, and the first business time segment is mapped to the time period to obtain the mapping relationship between the first business time segment and the time period. Based on the left-to-right order, the obtained mapping relationship is used to form a matrix representation of the time segment, thus obtaining the second business time segment; The second business time segment is input into a business time generation model pre-built based on the Transformer architecture to generate candidate business times for the target POI. The business time generation model is trained by the first business time sample data and the second business time sample data and the corresponding loss function. The first business time sample data is POIs containing business time, and the second business time sample data is POIs not containing business time and business time verified correctly by the business time discrimination model. Based on a pre-determined business hours discrimination model, the candidate business hours are discriminated, and a label for the candidate business hours is determined. In response to determining that the label is a second label, the candidate business hours are determined as the business hours of the target POI.
2. The method according to claim 1, wherein, The step of obtaining the time information of the target point of interest (POI) from web search results includes: Construct a combination of search keywords for POIs; Based on the combination of search keywords, the target POI is searched in the web search results to obtain the time information of multiple target POIs; From the time information of the multiple target POIs, obtain the time information that includes business hours.
3. The method of claim 1, wherein, The business hours discrimination model is trained based on negative sample training data and its corresponding first label and loss function, and positive sample training data and its corresponding second label and loss function. The negative sample training data consists of sample POIs with existing business hours, and the existing business hours are inconsistent with the generated business hours, or the negative sample training data consists of sample POIs that do not contain business hours. The positive sample training data consists of sample POIs with existing business hours, and the existing business hours are consistent with the generated business hours.
4. The method of claim 1, wherein, The method further includes: In response to determining that the label is the first label, the business hours and the first label are added to the negative sample training data of the business hours discrimination model.
5. The method of claim 1, wherein, The method further includes: Based on the intelligent telephone mode, the candidate business hours are inquired by telephone to obtain the inquiry results; In response to determining that the business hours in the query results match the candidate business hours, the candidate business hours are determined as the business hours of the target POI.
6. The method of claim 5, wherein, After determining the candidate business hours as the business hours of the target POI, the method further includes: Determine the second tag corresponding to the business hours of the target POI; The business hours and corresponding labels are added to the positive sample training data of the business hours discrimination model.
7. A time information determining device, the device comprising: The acquisition module is used to obtain the time information of the target point of interest (POI) from web search results. A generation module is used to: obtain a first business time segment from the time information, wherein the first business time segment is a time information segment obtained from web page text; determine the business time period of the target POI and segment the business time period in units of days; for each day, divide the day into different time periods, wherein the time periods include start time and end time; perform structural processing on the first business time segment based on the start time and end time of different time periods of each day, and map the first business time segment to the time periods to obtain the mapping relationship between the first business time segment and the time periods; based on the left-to-right order, construct a matrix representation of the time segments to obtain a second business time segment; input the second business time segment into a business time generation model pre-built based on the Transformer architecture to generate candidate business times for the target POI, wherein the business time generation model is trained by the first business time sample data and the second business time sample data and the corresponding loss function, wherein the first business time sample data is POI containing business time, and the second business time sample data is POI not containing business time and business time verified correctly by the business time discrimination model; The determination module is used to determine the candidate business hours based on a pre-determined business hours discrimination model, determine the label of the candidate business hours, and in response to determining the label as a second label, determine the candidate business hours as the business hours of the target POI.
8. The apparatus according to claim 7, wherein, The acquisition module is used for: Construct a combination of search keywords for POIs; Based on the combination of search keywords, the target POI is searched in the web search results to obtain the time information of multiple target POIs; From the time information of the multiple target POIs, obtain the time information that includes business hours.
9. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.
10. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.
11. A computer product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-6.