Method and system for providing search result reflecting place-related user's intention

Abstract

To provide a method for providing a search result reflecting a place-related user's intention to be executed by at least one processor.SOLUTION: A method for providing a search result includes the steps of: receiving a search query (query) from a user terminal; extracting a plurality of POI (Point of Interest) data sets related to the search query; and generating a search-result page based on the plurality of POI data sets.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a method and system for providing search results that reflect the intentions of location-related users, and more specifically, to a method and system for generating a search result page based on a plurality of POI (Point of Interest) datasets related to a search query.

Background Art

[0002] In the information age, search engines are positioned as essential tools for information retrieval. Search engines play a role in assisting users to effectively find the desired information on the web, and various search technologies have been attempted to provide information highly relevant to a search query.

[0003] Nevertheless, conventional search technologies, in providing search results, present the text, images, etc. included in the search results in a uniform form, regardless of the user's intention related to the search query or the specific theme the user is trying to investigate, and provide them to the user. As a result, conventional search technologies have limitations in providing search results that match the user's search intention, and there is a problem that additional effort is required, such as selecting the retrieved content and re-checking the full text of the content linked thereto, in order for the user to obtain the information they seek.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] The present disclosure provides a method and system (apparatus) for providing search results that reflect the intentions of location-related users in order to solve the above problems.

Means for Solving the Problems

[0006] The present disclosure can be realized by various methods including a method, an apparatus (system), or a computer program stored in a readable storage medium.

[0007] According to an embodiment of the present disclosure, a method for providing search results that reflect the intentions of location-related users, executed by at least one processor, includes receiving a search query from a user terminal, collecting a plurality of POI (Point of Interest) datasets related to the search query, and generating a search result page based on the plurality of POI datasets.

[0008] A computer-readable computer program for executing the above method by a computer according to an embodiment of the present disclosure is provided.

[0009] According to an embodiment of the present disclosure, an information processing system includes a communication module, a memory, and at least one processor coupled to the memory and configured to execute at least one computer-readable program included in the memory. The at least one program includes instructions for receiving a search query from a user terminal, extracting a plurality of POI datasets related to the search query, and generating a search result page based on the plurality of POI datasets.

Advantages of the Invention

[0010] According to various embodiments of the present disclosure, in generating a search result page, a plurality of POI datasets including information or data related to the location related to the search query can be utilized. Thereby, a search result page reflecting the search intention of the user related to a specific location can be provided to the user.

[0011] According to various embodiments of the present disclosure, a plurality of POI data sets related to a search query may include a plurality of review image data sets and review text data sets related to a location related to the search query. Thereby, by utilizing review data of a location related to a search query, which is a new data source, richer search results related to a location intended by a user can be provided.

[0012] According to various embodiments of the present disclosure, a machine learning model may be utilized in extracting a plurality of POI data sets related to a search query. Thereby, a POI data set more suitable for the search query can be effectively extracted.

[0013] According to various embodiments of the present disclosure, a plurality of snippets are extracted from a plurality of POI data sets, and a search result page may be generated to include a visual effect of highlighting at least a part of the plurality of snippets related to the search query. Thereby, a user can quickly confirm information related to a location to be searched without additional effort from the extracted snippets, and the reliability of the search results can be enhanced.

[0014] According to various embodiments of the present disclosure, a search result page includes a keyword filter catalog, and the search result page may be regenerated based on a selected keyword filter. Thereby, a search result page customized according to a user's keyword filter selection can be provided to the user.

[0015] The effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned are clearly understandable to a person having ordinary knowledge in the technical field to which the present disclosure belongs (referred to as "ordinary technician") from the description of the claims.

[0016] Embodiments of the present disclosure will be described with reference to the accompanying drawings described below. Note that similar reference numerals indicate similar elements, but are not limited thereto.

Brief Description of the Drawings

[0017]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

DETAILED DESCRIPTION OF THE INVENTION

[0018] Hereinafter, specific details for the implementation of the present disclosure will be described in detail with reference to the accompanying drawings. In the following description, when there is a risk of unnecessarily obscuring the gist of the present disclosure, specific descriptions of well-known functions and configurations will be omitted.

[0019] In the accompanying drawings, the same or corresponding components are given the same reference numerals. Also, in the description of the following embodiments, the description of the same or corresponding components may be omitted to avoid duplication. However, even if the description of a component is omitted, it is not intended that such a component is not included in a certain embodiment.

[0020] The advantages and features of the disclosed embodiments, as well as the methods for achieving them, will become apparent by referring to the embodiments described hereinafter together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below and can be realized in various different forms. Merely, these embodiments are provided to make the present disclosure complete and to enable those of ordinary skill in the art to accurately recognize the category of the invention.

[0021] The terms used in this specification will be briefly explained, and the disclosed embodiments will be specifically described. The terms used in this specification are selected as general terms that are currently widely used as much as possible while considering the functions in the present disclosure. However, this may change depending on the intentions or precedents of those skilled in the relevant fields, the emergence of new technologies, etc. Also, in certain cases, there are terms arbitrarily selected by the applicant, and in such cases, the meaning thereof will be described in detail in the description part of the invention. Therefore, the terms used in the present disclosure should be defined based not only on the name of the terms but also on the meaning they have and the overall content of the present disclosure.

[0022] As used herein, the singular forms also include the plural forms unless the context clearly dictates otherwise. Also, the plural forms include the singular form unless the context clearly dictates otherwise. Throughout the specification, when a portion is said to include a certain component, this means that it may further include other components rather than excluding other components unless otherwise stated.

[0023] Also, the terms "module" or "section" used in the specification mean software or hardware components, and the "module" or "section" performs some role. Note that the "module" or "section" is not limited to the meaning of software or hardware. The "module" or "section" may be configured to be in a storage medium that can be addressed, or may be configured to reproduce one or more processors. Thus, by way of example, the "module" or "section" may include components such as software components, object-oriented software components, class components, and task components, and at least one of processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, or variables. The components and the "module" or "section" may be combined with a smaller number of further components and "modules" or "sections" provided therein, or may be further separated into additional components and "modules" or "sections".

[0024] According to an embodiment of the present disclosure, a "module" or "unit" can be implemented as a processor and a memory. The "processor" should be broadly construed to include a general-purpose processor, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a controller, a microcontroller, a state machine, and the like. In some environments, the "processor" may also refer to a custom semiconductor (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), and the like. The "processor" may refer to a combination of processing devices, such as, for example, a combination of a DSP and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors coupled with a DSP core, or a combination of any other such configuration. Also, the "memory" should be broadly construed to include any electronic component capable of storing electronic information. The "memory" may refer to various types of processor-readable media, such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable-programmable read-only memory (EPROM), electrically erasable PROM (EEPROM), flash memory, magnetic or optical data storage devices, registers, and the like. If the processor can read information from and / or write information to the memory, it can be said that the memory is in an electronic communication state with the processor. The memory integrated into the processor is in an electronic communication state with the processor.

[0025] In the present disclosure, a "system" may include, but is not limited to, at least one device of a server device and a cloud device. For example, the system may be composed of one or more server devices. As another example, the system may be composed of one or more cloud devices. As yet another example, the system may be composed and operated with both a server device and a cloud device.

[0026] In the present disclosure, "each of a plurality of As" or "each of the plurality of As" may refer to each of all the components included in the plurality of As, or may refer to each of some of the components included in the plurality of As. For example, each of the plurality of POI datasets may refer to each of all the POI datasets included in the plurality of POI datasets, or may refer to each of some of the POI datasets included in the plurality of POI datasets.

[0027] In the present disclosure, "POI (Point of Interest) data" may refer to data related to a location that a user wants to search for or is interested in. For example, the POI data may include information or data (such as review image data or review text data) regarding opinions, evaluations, emotional expressions, reviews, etc. related to the location uploaded by users who have visited a specific location.

[0028] In the present disclosure, a "snippet" may refer to a part of the content or document (such as a web page, document, text, or video content, etc.) retrieved by a search query and published as a search result. The snippet may be automatically generated based on the content or document retrieved by the search query. Specifically, the snippet may be generated to highlight a part of the content or document that is highly relevant to the search query or search keyword, or to provide a preview or summary thereof. For example, in the case of a document retrieved by a search query, the snippet may be displayed in a form that extracts a part of the document text where the search keyword included in the search query is matched and highlights the keyword in the corresponding part, or may be displayed in the form of a representative description or summary text prepared in advance for the corresponding document. As another example, the snippet may be generated to provide a part of an image related to the search query.

[0029] FIG. 1 shows an example in which a search result page 120 related to a location generated in response to a search query input by user 100 via user terminal 110 is provided according to an embodiment of the present disclosure. For example, user 100 may utilize a search service to perform a location-related search by executing a search application on user terminal (e.g., smartphone, etc.) 110.

[0030] According to one embodiment, user 100 may receive the provision of a search result page related to a location generated by an information processing system (e.g., a search server) based on the search query by inputting a search query related to the location into search term input section 130 output on the search application. For example, user 100 may input a search query related to a specific location (e.g., "Jiangnan Terrace Gourmet") into search term input section 130, and correspondingly receive the provision of search result page 120 including information about locations related to the search query (e.g., "Cafeteria A", "Cafeteria B", etc.).

[0031] According to one embodiment, user terminal 110 may display search result page 120 including a plurality of POI data sets searched based on the search query input by user 100. For example, when user 100 inputs "Jiangnan Terrace Gourmet" as the search query, search result page 120 may include a plurality of POI data sets related to locations such as "Cafeteria A" and "Cafeteria B" that are related to "Jiangnan Terrace Gourmet". Here, search result page 120 is configured to arrange the plurality of POI data sets in order according to a ranking determined by a specific criterion, and may preferentially include the POI data sets corresponding to the upper rankings in the upper regions.

[0032] According to one embodiment, the POI dataset 140 may include a review image dataset 142 and a review text dataset 144. The review image dataset 142 may include a plurality of review images taken and uploaded by visitors to a location related to the search query. The review text dataset 144 may include a plurality of review texts created by visitors to a location related to the search query. In this case, the plurality of review image data included in the review image dataset 142 may be arranged in a ranking order determined by specific criteria. Similarly, the plurality of review text data included in the review text dataset 144 may also be arranged in a ranking order determined by specific criteria.

[0033] According to one embodiment, the search result page 120 may include the POI dataset 140 in a snippet form. For example, snippets may be extracted from each of the plurality of image data included in the review image dataset 142, and the extracted plurality of snippets may be arranged in a ranking order determined by the specific criteria described above. In this case, the snippet extracted from each of the plurality of image data may be a part of the corresponding image. Similarly, snippets may be extracted from each of the plurality of text data included in the review text dataset 144, and the extracted plurality of snippets may be arranged in a ranking order determined by the aforementioned specific criteria. In this case, the snippet extracted from each of the plurality of text data may be a part of the text included in the corresponding text. Also, when the search result page 120 includes the POI dataset 140 in a snippet form, a visual effect of emphasizing at least a part of the plurality of snippets related to the search query may be applied. For example, when the user 100 enters "Gangnam Terrace Gourmet" as a search query, "Terrace" included in the snippet extracted from the review text data of the POI dataset 140 on the search result page 120 may be displayed in bold.

[0034] According to one embodiment, an information processing system that provides a search result page 120 may receive a search query from a user terminal 110, extract a plurality of POI data sets related to the search query, and then generate and provide a search result page 120 including the extracted plurality of POI data sets. To that end, the information processing system may use a machine learning model to map a label to each of a plurality of candidate review data sets related to the search query, calculate the similarity between the label and the search query, and determine at least a part of the plurality of candidate review text data sets as the plurality of POI data sets. Also, a plurality of candidate data sets may be used for extraction of the search query and the POI data sets (review text data sets related to the search query, review image data sets related to the search query). The plurality of candidate data sets may include a candidate review text data set and a candidate review image data set. The plurality of candidate data sets may be generated in association with each other. For example, the candidate review text data and the candidate review image data may be created and uploaded together by a user. Details regarding the extraction of the plurality of POI data sets based on the search query and the method of generating the search result page will be described later with reference to FIGS. 4 to 7.

[0035] According to one embodiment, an information processing system that provides a search result page 120 may generate the search result page 120 based on a plurality of POI data sets related to the search query. To that end, the information processing system may determine a ranking for each of the plurality of POI data sets and may determine a ranking for each of the plurality of POI data (e.g., a plurality of review text data and a plurality of review image data) included in the plurality of POI data sets. Details regarding this will be described later with reference to FIG. 4.

[0036] According to the previous configuration, in generating the search result page 120, a plurality of POI data sets including information or data related to locations related to the search query can be utilized. Thereby, the search result page 120 reflecting the user's search intention related to a specific location can be provided to the user 100.

[0037] Also, the POI data set 140 related to the search query may include a review image data set 142 related to the search query and a review text data set 144 related to the search query. Thereby, by utilizing the review data of the location related to the search query, which is a new data source, richer search results can be provided to the user.

[0038] Furthermore, a plurality of snippets are extracted from the plurality of POI data sets, and the search result page 120 can be generated so as to include a visual effect that emphasizes at least a part of the plurality of snippets related to the search query. Thereby, the user 100 can quickly confirm information about the location to be searched without additional effort from the extracted snippets, and the reliability of the search results can be enhanced.

[0039] FIG. 2 is a schematic diagram showing a configuration in which an information processing system 230 is connected to be communicable with a plurality of user terminals 210_1, 210_2, 210_3 in order to provide a search result page in response to an input of a location-related search query according to an embodiment of the present disclosure. As shown in the figure, the plurality of user terminals 210_1, 210_2, 210_3 can be connected to an information processing system 230 that can provide a search result page via a network 220. Here, the plurality of user terminals 210_1, 210_2, 210_3 may include terminals of users who receive an input of a search query from the user and are provided with a search result page generated in association with the search query.

[0040] In one embodiment, the information processing system 230 may include one or more server devices and / or databases that can store, provide, and execute computer-executable programs (e.g., downloadable applications) and data related to generating search result pages based on search queries, or one or more distributed computing devices and / or distributed databases based on cloud computing services.

[0041] The search result pages provided by the information processing system 230 can be provided to users by, for example, a search service application, a web browser, or a web browser extension installed on each of the plurality of user terminals 210_1, 210_2, 210_3. For example, the information processing system 230 can provide information corresponding to search requests received from the user terminals 210_1, 210_2, 210_3 or execute corresponding processing by, for example, a search service application.

[0042] The plurality of user terminals 210_1, 210_2, 210_3 can communicate with the information processing system 230 via the network 220. The network 220 can be configured to enable communication between the plurality of user terminals 210_1, 210_2, 210_3 and the information processing system 230. Depending on the installation environment, the network 220 can be composed of, for example, a wired network such as Ethernet (registered trademark), a wired home network (Power Line Communication), a telephone line communication device, and RS serial communication, a mobile communication network, a WLAN (Wireless LAN), Wi-Fi (registered trademark), Bluetooth (registered trademark), and ZigBee (registered trademark), or a combination thereof. The communication method is not limited, and it includes not only communication methods that utilize communication networks that the network 220 can include (as an example, a mobile communication network, a wired Internet, a wireless Internet, a broadcast network, a satellite network, etc.), but also short-range wireless communication between the user terminals 210_1, 210_2, 210_3.

[0043] In FIG. 2, the mobile phone terminal 210_1, the tablet terminal 210_2, and the PC terminal 210_3 are shown as examples of user terminals, but are not limited thereto. The user terminals 210_1, 210_2, and 210_3 can be any computing device capable of wired and / or wireless communication and on which a search service application or a web browser, etc. can be installed and executed. For example, the user terminal can include an AI speaker, a smartphone, a mobile phone, a navigation device, a computer, a notebook, a digital broadcast terminal, a PDA (Personal Digital Assistants), a PMP (Portable Multimedia Player), a tablet PC, a game console, a wearable device, an IoT (internet of things) device, a VR (virtual reality) device, an AR (augmented reality) device, a set-top box, etc. Also, in FIG. 2, three user terminals 210_1, 210_2, and 210_3 are illustrated as communicating with the information processing system 230 via the network 220, but are not limited thereto, and a different number of user terminals can also be configured to communicate with the information processing system 230 via the network (220).

[0044] FIG. 3 is a block diagram showing the internal configurations of user terminal 210 and information processing system 230 according to an embodiment of the present disclosure. User terminal 210 can refer to any computing device on which applications, web browsers, etc. can be executed and which can perform wired / wireless communication. For example, it may include mobile phone terminal 210_1, tablet terminal 210_2, PC terminal 210_3, etc. of FIG. 2. As shown in the figure, user terminal 210 may include memory 312, processor 314, communication module 316, and input / output interface 318. Similarly, information processing system 230 may include memory 332, processor 334, communication module 336, and input / output interface 338. As shown in FIG. 3, user terminal 210 and information processing system 230 may be configured to communicate information and / or data via network 220 using respective communication modules 316, 336. Also, input / output device 320 may be configured to input information and / or data to user terminal 210 or output information and / or data generated from user terminal 210 via input / output interface 318.

[0045] Memories 312 and 332 may include any non-transitory computer-readable recording medium. According to one embodiment, memories 312 and 332 may include non-volatile mass storage devices such as ROM (read only memory), disk drive, SSD (solid state drive), flash memory, etc. As another example, non-volatile mass storage devices such as ROM, SSD, flash memory, disk drive, etc. may be included in user terminal 210 or information processing system 230 as separate permanent storage devices distinct from the memory. Also, an operating system and at least one program code (for example, code related to a location search service, etc.) may be stored in memories 312 and 332.

[0046] Such software components can be loaded from a computer-readable recording medium separate from memories 312 and 332. Such a separate computer-readable recording medium may include a recording medium directly connectable to such user terminal 210 and information processing system 230. For example, it may include computer-readable recording media such as a floppy (registered trademark) drive, a disk, a tape, a DVD / CD-ROM drive, a memory card, etc. As another example, the software components may be loaded into memories 312 and 332 via communication modules 316 and 336 instead of a computer-readable recording medium. For example, at least one program may be loaded into memories 312 and 332 based on a computer program (e.g., a program for location search) installed by a file provided by a file distribution system that distributes developer or application installation files via network 220.

[0047] Processors 314 and 334 may be configured to process instructions of a computer program by performing basic arithmetic, logic, and input / output operations. The instructions may be provided to processors 314 and 334 by memories 312 and 332 or communication modules 316 and 336. For example, processors 314 and 334 may be configured to execute received instructions by program code stored in a recording device such as memories 312 and 332.

[0048] The communication modules 316 and 336 can provide a configuration or function for the user terminal 210 and the information processing system 230 to communicate with each other via the network 220, and the user terminal 210 and / or the information processing system 230 can provide a configuration or function for communicating with other user terminals or other systems (for example, separate cloud systems, etc.). As an example, a request or data (for example, a request for providing search results corresponding to a search query, etc.) generated by a program code stored in a recording device such as the memory 312 by the processor 314 of the user terminal 210 can be transmitted to the information processing system 230 via the network 220 under the control of the communication module 316. Conversely, a control signal or instruction provided under the control of the processor 334 of the information processing system 230 can be received by the user terminal 210 via the communication module 316 of the user terminal 210 through the communication module 336 and the network 220.

[0049] The input / output interface 318 may be means for interfacing with the input / output device 320. As an example, the input device may include devices such as a camera including an audio sensor and / or an image sensor, a keyboard, a microphone, a mouse, etc., and the output device may include devices such as a display, a speaker, a haptic feedback device, etc. As another example, the input / output interface 318 may be means for interfacing with a device in which the configuration or function for performing input and output is integrated into one, such as a touch screen. For example, in the processor 314 of the user terminal 210 processing the instructions of the computer program loaded in the memory 312, a service screen configured using the information and / or data provided by the information processing system 230 or other user terminals may be displayed on the display via the input / output interface 318. In FIG. 3, the input / output device 320 is shown not to be included in the user terminal 210, but is not limited thereto and may be configured as one device with the user terminal 210. Also, the input / output interface 338 of the information processing system 230 may be means for interfacing with a device (not shown) that is connected to the information processing system 230 or for input or output that the information processing system 230 may include. In FIG. 3, the input / output interfaces 318, 338 are shown as elements configured separately from the processors 314, 334, but are not limited thereto and may be configured to be included in the processors 314, 334.

[0050] The user terminal 210 and the information processing system 230 may include more components than those shown in FIG. 3. However, it is not necessary to clearly show most of the conventional components. In one embodiment, the user terminal 210 may be implemented to include at least a part of the input / output devices 320 described above. Further, the user terminal 210 may further include other components such as a transceiver, a GPS (Global Positioning System) module, a camera, various sensors, a database, etc. For example, when the user terminal 210 is a smartphone, it can include components generally included in a smartphone. For example, various components such as an acceleration sensor, a gyro sensor, a microphone module, a camera module, various physical buttons, buttons using a touch panel, an input / output port, and a vibrator for vibration can be implemented to be further included in the user terminal 210.

[0051] While a program such as a search service application is operating, the processor 314 can receive text, images, videos, audio, and / or operations input or selected via input devices including a touch screen, a keyboard, an audio sensor, and / or a camera including an image sensor, a microphone, etc., connected to the input / output interface 318, and can store the received text, images, videos, audio, and / or operations, etc. in the memory 312 or provide them to the information processing system 230 via the communication module 316 and the network 220.

[0052] The processor 314 of the user terminal 210 may be configured to manage, process, and / or store information and / or data received from the input / output device 320, other user terminals, the information processing system 230, and / or a plurality of external systems. The information and / or data processed by the processor 314 may be provided to the information processing system 230 via the communication module 316 and the network 220. The processor 314 of the user terminal 210 may output by transferring information and / or data to the input / output device 320 via the input / output interface 318. For example, the received information and / or data may be displayed on the screen of the user terminal 210.

[0053] The processor 334 of the information processing system 230 may be configured to manage, process, and / or store information and / or data received from a plurality of user terminals 210 and / or a plurality of external systems. The information and / or data processed by the processor 334 may be provided to the user terminal 210 via the communication module 336 and the network 220. In one embodiment, the processor 334 may execute instructions to extract a plurality of POI data sets related to a search query received from the user terminal 210 and generate a search result page based on the plurality of POI data sets.

[0054] FIG. 4 is a block diagram showing an internal configuration of a processor of an information processing system according to an embodiment of the present disclosure. As shown in FIG. 4, the processor 334 may include a similarity calculation unit 410, a data extraction unit 420, a ranking determination unit 430, and a search result generation unit 440. In FIG. 4, each component of the processor 334 shows a functional element that is functionally divided, and a plurality of components may be realized in a form integrated with each other in an actual physical environment. Alternatively, each component of the processor 334 may be realized separately from each other in an actual physical environment.

[0055] According to one embodiment, a plurality of candidate datasets may be used for extracting a search query and a POI dataset (a review text dataset related to the search query, a review image dataset related to the search query). The plurality of candidate datasets may include a candidate review text dataset and a candidate review image dataset. The plurality of candidate datasets may be generated in association with each other. For example, the candidate review text data and the candidate review image data may be created and uploaded together by a user. According to one embodiment, the similarity calculation unit 410 may calculate the similarity between the plurality of candidate datasets and the search query in order to extract a plurality of POI datasets related to the search query. To that end, the similarity calculation unit 410 may map a label to each of the plurality of candidate datasets (for example, the candidate review text dataset and the candidate review image dataset), and may include at least one machine learning model for performing the label mapping. In this case, the label mapping and the similarity calculation for each of the candidate review text dataset and the candidate review image dataset may be performed independently of each other. For details of the detailed configuration of the machine learning model and the details of mapping a label to each of the plurality of candidate datasets using the machine learning model, reference will be made to FIGS. 5 to 7 and described later.

[0056] According to one embodiment, the data extraction unit 420 may extract a plurality of POI data sets related to the search query. The data extraction unit 420 may determine at least a part of the plurality of candidate data sets as a plurality of POI data sets related to the search query based on the similarity calculated by the similarity calculation unit 410. For example, the data extraction unit 420 may determine, as a POI data set, a candidate data set whose similarity calculated by the similarity calculation unit 410 is equal to or greater than a predetermined threshold value. In this case, the threshold value may be adjusted and predetermined according to the distribution of similarity, the intention of the user, and the like. Additionally or alternatively, the data extraction unit 420 may extract a POI data set related to the search query in consideration of the data amount of the candidate data set, the number of user clicks on the candidate data set, and the like. For example, when the number of candidate data sets with a similarity equal to or greater than the threshold value is a considerable number exceeding a predetermined value, the data extraction unit 420 may preferentially determine, as a POI data set related to the search query, a candidate data set including a large amount of data or a candidate data set including data searched by a large number of users. In this case, in the extraction of the POI data set, the extraction of the review text data set related to the search result and the extraction of the review image data set related to the search result may be executed independently of each other.

[0057] According to one embodiment, the ranking determination unit 430 may determine the ranking of a plurality of POI data sets extracted by the data extraction unit 420. In one embodiment, the ranking determination unit 430 may determine the ranking for each of the plurality of POI data sets. In this case, the ranking determination unit 430 may determine the ranking for each of the plurality of POI data sets based on at least one of the number of POI data, the number of occurrences of a keyword, or the appearance rate of a keyword. For example, the ranking determination unit 430 may determine that the higher the number of POI data included in a POI data set, the higher the number of occurrences of a keyword related to the search query from the POI data included in the POI data set, or the higher the ratio of the POI data in which the keyword related to the search query appears to the total POI data, the higher the ranking. In addition, the ranking determination unit 430 may determine the ranking of the POI data set by weighting and adopting the above-described ranking determination criteria in order to provide an optimal search result. The criteria for determining the ranking for each of the plurality of POI data sets are not limited to the above-described criteria, and the ranking determination unit 430 may determine the ranking for each of the plurality of POI data sets based on other criteria other than the above-described criteria.

[0058] According to one embodiment, the ranking determination unit 430 may determine the ranking for each of the plurality of POI data included in the plurality of POI data sets. In one embodiment, the ranking determination unit 430 may determine the ranking for each of the plurality of review image data included in the review image data set related to the search query. In this case, the ranking determination unit 430 may determine the ranking for each of the plurality of review image data based on the similarity or the recency with the search query. For example, the ranking determination unit 430 may determine that the higher the similarity between the POI data and the search query, or the more recent the POI data generation date, the higher the ranking of the POI data. To that end, the ranking determination unit 430 may use the similarity calculated by the similarity calculation unit 410.

[0059] In one embodiment, the ranking determination unit 430 can determine rankings for each of a plurality of review text data included in the review text data set related to the search query. In this case, the ranking determination unit 430 can determine rankings for each of the plurality of review text data based on at least one of the recency of the review text data, the number of occurrences of relevant keywords in the review text data, or the amount of information included in the text data. For example, the ranking determination unit 430 can determine a higher ranking as the generation date of the review text data is more recent, as there are more sentences containing keywords related to the search query in the review text data, or as there is more text containing significant information (e.g., menu items sold at a specific location, the atmosphere of a specific location, etc.) in the review text data.

[0060] Also, the ranking determination unit 430 can independently determine rankings for each of the plurality of review text data and rankings for each of the plurality of review image data. For example, even if image data and text data are combined and present in any review data, rankings can be independently determined for each of the review image data and the review text data.

[0061] The ranking determination unit 430 can weight and adopt the aforementioned ranking criteria in order to provide optimal search results. Note that the criteria for determining rankings for each of the plurality of POI data are not limited to the aforementioned criteria, and the ranking determination unit 430 can determine rankings for each of the plurality of POI data sets based on other criteria other than the aforementioned criteria.

[0062] According to an embodiment, the search result generation unit 440 may generate a search result page including the extracted POI data set. In this case, the search result generation unit 440 may extract a plurality of snippets from each of the extracted POI data sets. For example, the search result generation unit 440 may extract a specific part (e.g., the central part, etc.) of the review image data related to the search query as a snippet for each POI data set. As another example, the search result generation unit 440 may extract a specific sentence of the review text data related to the search query as a snippet for each POI data set. In this case, the search result generation unit 440 may apply a visual effect of emphasizing at least a part of the plurality of snippets related to the search query. For example, the search result generation unit 440 may apply a visual effect of displaying the text snippet related to the search query in bold. Also, the search result generation unit 440 may generate a search result page by arranging a plurality of snippets according to the ranking determined by the ranking determination unit 430. For example, the search result generation unit 440 may extract a snippet set from the POI data set and arrange the snippet set vertically according to the ranking of each of the plurality of POI data sets determined by the ranking determination unit 430. Then, the search result generation unit 440 may arrange the snippets included in the snippet set horizontally according to the ranking of each of the plurality of POI data determined by the ranking determination unit 430.

[0063] In FIG. 4, the internal configuration of the processor 334 is shown divided into a similarity calculation unit 410, a data extraction unit 420, a ranking determination unit 430, and a search result generation unit 440, but is not limited thereto, and some configurations may be omitted or other configurations may be added.

[0064] FIG. 5 is a diagram showing an example of a method for training a machine learning model 530 according to an embodiment of the present disclosure. The machine learning model 530 in FIG. 5 can be a model that maps labels to candidate review image data when a search query related to the context of a location is received. In this case, the machine learning model 530 can be configured to include a language embedding model (e.g., Simsce, Bert, etc.) 532 and an image model (e.g., CLIP, etc.) 534.

[0065] According to one embodiment, the machine learning model 530 can be trained to cluster a training image dataset into a plurality of clusters and map a label to each of the plurality of clusters. For example, the machine learning model 530 can receive a context candidate 510 and a training review image dataset 520. In this case, the context candidate 510 can refer to candidate labels that can be defined as the context of the image data. Then, the machine learning model 530 can combine the embedding vector of the keyword extracted from the context candidate 510 and the embedding vector of the image extracted from the training review image dataset 520, and cluster the training review image dataset 520 into a plurality of clusters based on this. Then, the machine learning model 530 can assign a context index 540 corresponding to each of the plurality of clusters. In this case, the context index 540 can refer to a label related to the context.

[0066] According to one embodiment, the machine learning model 530 can utilize the context index 540 as reference data in the label mapping of candidate data. For example, when labels for the context of a candidate review image set are required, the machine learning model 530 calculates the distance values between each of the candidate review image sets and each of the clusters with the context index 540 assigned, and can map the context index 540 of the cluster whose distance value is below the threshold as the label of the candidate review image set. With the above configuration, the context of an image that is difficult to label can be labeled according to the reference of the context index 540. Also, in extracting review image data for a search query related to the context of a location, the reliability of the search results can be enhanced by utilizing the mapped labels.

[0067] FIG. 6 is a diagram showing an example of a method in which a machine learning model 620 maps labels to candidate review image data according to an embodiment of the present disclosure. The machine learning model 620 in FIG. 6 can be a model that maps labels to candidate review image data when a search query related to the name of a menu or food is received.

[0068] According to one embodiment, the machine learning model 620 may map labels to candidate review image data. In this case, the label may include the name of the food recognized from the image, the accuracy of the recognized food name, and / or the top-category label including the recognized food name. For example, the machine learning model 620 may receive an input of a food photo 610 containing "pasta" and map the corresponding label 630. In this case, the label 630 may include the name of the food recognized from the photo 610 (e.g., "Bongole"), the probability that the food included in the photo 610 is "Bongole" (99.8%), and / or the top-category label of the food (e.g., "pasta"). As another example, the machine learning model 620 may receive an input of a photo 612 without food and map the corresponding label 632. Similarly, the label 632 may include the name of the food recognized from the photo 612 (e.g., "ice cream"), the probability that the food included in the photo 612 is "ice cream" (11.3%), and the top-category label of the food (e.g., no corresponding category). The labels mapped in this way can be utilized to extract a review image data set related to the search query by the machine learning model 620 and exclude candidate review image data sets unrelated to the search query. Also, by including the top-category label in the label output to the machine learning model 620, even if the search item does not exactly match the intention of the search query in the label, candidate image data sets belonging to the same / similar category can be included in the image data related to the search query.

[0069] FIG. 7 is a diagram showing an example of a method in which a machine learning model maps labels to candidate review text data according to an embodiment of the present disclosure. The machine learning model 720 in FIG. 7 may be a model that maps labels to a candidate review text data set. In one embodiment, the machine learning model 720 may be a BERT-CRF model including a BERT model 724 and a CFR model 726.

[0070] According to one embodiment, the preprocessing unit 722 may perform preprocessing to convert the candidate text data set into data in a form suitable for inputting into the machine learning model 720. For example, the preprocessing unit 722 may split the candidate text data set into text data 710 in units of sentences, such as "A dog-friendly café accidentally found in the neighborhood", and may further tokenize the text data 710 into a plurality of tokens.

[0071] According to one embodiment, the BERT model 724 of the machine learning model 720 may extract embedding vectors from each of the plurality of tokens output by the preprocessing unit 722. In this case, the embedding vectors may include the input token sequence and semantic information regarding each token.

[0072] According to one embodiment, the CFR model 726 of the machine learning model 720 may map labels to each of the plurality of candidate review text data sets based on the embedding vectors extracted by the BERT model 724. The CFR model 726 may map labels considering the dependency relationships between labels within a token sequence including a plurality of tokens. For example, the CFR model 726 may extract labeled text data 730 by mapping "dog-friendly café" to the text data 710, which is "A dog-friendly café accidentally found in the neighborhood".

[0073] As described above, the machine learning model 720 can map labels to a plurality of candidate review text data sets, and the mapped labels can be utilized to search for a review text data set related to the location intended by the search query.

[0074] FIG. 8 is a diagram showing an example of a search result page generated according to an embodiment of the present disclosure. The first screen 810 is an illustration of a search result page output when a search query related to the context of a location is input. For example, when searching for "Kounan Terrace Gourmet", the review image dataset 812 and the review text dataset 814 related to "Terrace" can be arranged in the determined ranking order in snippet form. In this case, the user can further check the lower-ranked image data and text data by swiping the first screen 810 to the left. In one embodiment, the review image dataset 812 is extracted from a group of review data candidates including keyword data related to a plurality of locations (such as a cafeteria, a café, etc.) with a large number of times (i.e., query count) used as at least a part of the search query, with reference to the review text written together with the review images included in the group of review data candidates. Also, the review text dataset 814 may include context keywords related to the location and may include non-negative and significant review text. Further, when the review text dataset 814 includes a keyword related to the search query, a visual effect of emphasizing the keyword may be further reflected.

[0075] The second screen 820 is an example of a search result page output when a search query related to the food menu of a location is entered. For example, when the user enters "John Jadon Pasta" as a search query, the review image dataset 822 and review text dataset 824 related to pasta can be arranged in the determined ranking order in snippet form. In this case, the user can further check the lower-ranked image data and text data by swiping the second screen 820 to the left. In one embodiment, the review image dataset 822 can be an image related to the menu extracted by a machine learning model. Also, the review text dataset 824 can include the menu name and may include non-negative and significant review text. Further, when the review text dataset 824 includes keywords related to the search query, a visual effect of emphasizing the keywords can be further reflected.

[0076] FIG. 9 is a diagram showing an example of a search result page and keyword statistical data regenerated based on a keyword filter according to an embodiment of the present disclosure. The first screen 910 is an example of outputting a search result page regenerated based on a keyword filter. The search result page may include a keyword filter catalog 912, a plurality of POI dataset 916 that are search results, and a keyword filter application display area 918.

[0077] According to one embodiment, the keyword filter catalog 912 may include a plurality of keyword filters. In one embodiment, the keyword filter catalog 912 may include keyword filters related to a plurality of POI datasets. For example, the keyword filters in the keyword filter catalog 912 can be keywords extracted from the keyword statistical data of a plurality of locations related to the top-ranked POI dataset. As another example, the keyword filters in the keyword filter catalog 912 can be personalized keyword filters that reflect the user's keyword filter selection history.

[0078] According to an embodiment, a user can select a specific keyword filter 914 from the keyword filter catalog 912 for which the user wants to filter search results, and can receive the provision of filtered search results based on the selected keyword filter. For this purpose, an information processing system that provides a search result page receives a user input for selecting at least one keyword filter from the keyword filter catalog 912 from a user terminal, and can regenerate the search result page based on the keyword filter selected according to the user input.

[0079] According to an embodiment, an information processing system that provides a search result page can determine the ranking of each of a plurality of POI data sets based on the selected keyword filter, and can regenerate the search result page based on the determined ranking. In this case, the information processing system can determine the ranking based on keyword statistical data collected by visitors to a specific location. Details regarding this will be described later with reference to the second screen 920.

[0080] According to an embodiment, the keyword filter application display area 918 can display data related to the state in which the keyword filter is applied. For example, when the keyword filter is applied, the keyword filter application display area 918 can change the color of the area for display and can display the number of applied keyword filters together.

[0081] The second screen 920 is an example of outputting a keyword statistical data display area 922. A visitor to a specific location can select a keyword related to the location, and keyword statistical data can be collected based on the selected keyword. The collected keyword statistical data can be displayed in the form of a progress bar in the keyword statistical data display area 922 of a web page related to the location.

[0082] According to an embodiment, the POI dataset for a specific location is associated with the keyword statistical data of the specific location. When a search result page applying a keyword filter is generated, rankings for each of a plurality of POI datasets can be determined based on the keyword statistical data. For example, according to the keyword statistical data of "Canteen E", it can be confirmed that a large number of visitors have selected the keyword "There is a special menu". Thus, when the keyword filter 914 of "special menu" is selected from the keyword filter catalog 912, the information processing system can determine the POI set of "Canteen E" as the top ranking. In this case, when the information processing system regenerates the search result page based on the selected keyword filter, it can determine the ranking by adjusting the weights of the keyword selection count and the keyword selection rate in the keyword statistical data. With the above configuration, a search result page customized according to the user's selection of the keyword filter can be provided to the user.

[0083] FIG. 10 is a flowchart showing an example of a search result providing method 1000 that reflects the user's intention related to a location according to an embodiment of the present disclosure. In one embodiment, the method 1000 can be executed by at least one processor (e.g., processor 334) of the information processing system and / or at least one processor (e.g., processor 314) of the user terminal. Alternatively, each step of the method 1000 can be executed separately by at least one processor of the user terminal and at least one processor of the information processing system.

[0084] The method 1000 can be started by receiving a search query from the user terminal (S1010). Thereafter, the processor can collect a plurality of POI datasets related to the search query (S1020). In one embodiment, the plurality of POI datasets related to the search query can include a plurality of review image datasets related to the search query and a plurality of review text datasets related to the search query.

[0085] In one embodiment, the processor may use a first machine learning model to map labels to each of a plurality of candidate review image datasets, calculate a similarity between the labels and a search query, and determine at least a portion of the plurality of candidate review image datasets as a plurality of POI datasets based on the similarity. In this case, the plurality of candidate review image datasets may be generated in association with a plurality of candidate review text datasets.

[0086] In one embodiment, the first machine learning model may be trained to cluster a training image dataset into a plurality of clusters and map labels to each of the plurality of clusters. Further, the processor may use the first machine learning model to calculate a distance value between each of the plurality of candidate review image datasets and each of the plurality of clusters, and map the label of the cluster whose distance value is equal to or less than a threshold value to each of the plurality of candidate review image datasets. In this case, the label may include a top-level category label.

[0087] In one embodiment, the processor may use a second machine learning model to map labels to each of a plurality of candidate review text datasets, calculate a similarity between the labels and a search query, and determine at least a portion of the plurality of candidate review text datasets as a plurality of POI datasets based on the similarity.

[0088] In one embodiment, the processor may tokenize a plurality of candidate review text datasets into a plurality of tokens, use a second machine learning model to extract an embedding vector including semantic information from each of the plurality of tokens, and use the second machine learning model to map labels to each of the plurality of candidate review text datasets based on the embedding vectors.

[0089] Finally, the processor may generate a search result page based on a plurality of POI data sets (S1030). In one embodiment, the processor determines rankings for each of the plurality of POI data sets, determines rankings for each of the plurality of POI data included in the plurality of POI data sets, and may generate a search result page based on the rankings for each of the plurality of POI data sets and the rankings for each of the plurality of POI data.

[0090] In one embodiment, the processor may determine rankings for each of the plurality of POI data sets based on at least one of the number of POI data, the number of occurrences of keywords, or the keyword occurrence rate. Further, the processor may determine rankings for each of the plurality of review image data included in the review image data set related to the search query, and may determine rankings for each of the plurality of review text data included in the review text data set related to the search query. In this case, the step of determining rankings for each of the plurality of review text data and the step of determining rankings for each of the plurality of review image data may be executed independently of each other.

[0091] In one embodiment, the processor may determine rankings for each of the plurality of review image data based on the similarity or recency with the search query. Further, the processor may determine rankings for each of the plurality of review text data based on at least one of recency, the number of occurrences of keywords, or the amount of information included in the text data.

[0092] In one embodiment, the processor may extract a plurality of snippets from the plurality of POI data sets and generate a search result page including a visual effect of highlighting at least a part of the plurality of snippets related to the search query.

[0093] In one embodiment, the search result page may include a keyword filter catalog for filtering search results. Further, the processor may receive user input for selecting at least one keyword filter from the keyword filter catalog from the user terminal, and in response to the user input, may regenerate the search result page based on the selected keyword filter.

[0094] In one embodiment, the processor may determine the ranking of each of the plurality of POI data sets based on the selected keyword filter, and regenerate the search result page by sorting the plurality of POI data sets based on the determined ranking.

[0095] In one embodiment, each of the plurality of POI data sets may be associated with keyword statistical data including keyword selection results related to a specific location by visitors to the specific location. Further, the processor may determine the ranking of each of the plurality of POI data sets based on the selected keyword filter and the keyword statistical data. Furthermore, the keyword filter catalog may include keyword filters related to the plurality of POI data sets and personalized keyword filters.

[0096] The flowchart shown in FIG. 10 and the foregoing description are merely examples, and different realizations may be obtained in some embodiments. For example, in some embodiments, the procedures of each step may change, some steps may be repeatedly executed, some steps may be omitted, or some steps may be added.

[0097] The prior method can be provided as a computer program stored on a computer-readable recording medium for execution by a computer. The medium can be one that continuously stores a computer-executable program, or temporarily stores it for execution or download. Also, the medium can be various recording or storage means in a form combined with single or multiple hardware, but is not limited to a medium directly connected to any computer system, and can also exist distributed on a network. Examples of the medium include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto optical media such as floptical disks, and those configured to store program instruction words including ROM, RAM, flash memory, etc. Also, as examples of other media, there can be mentioned app stores that distribute applications, and recording media or storage media managed by sites, servers, etc. that supply or distribute various other software.

[0098] The methods, operations, or techniques of the present disclosure can be implemented by various means. For example, such techniques may be implemented by hardware, firmware, software, or combinations thereof. Those of ordinary skill in the art will understand that the various exemplary logical blocks, modules, circuits, and algorithm steps described in connection with the disclosure of this application can also be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate such interchangeability between hardware and software, various exemplary components, blocks, modules, circuits, and steps have been generally described above in terms of their functional aspects. Whether such functions are implemented as hardware or as software will vary depending on the particular application and design requirements imposed on the overall system. Those of ordinary skill in the art can implement the functions described in various ways for each particular application, but such implementations should not be construed as departing from the scope of the present disclosure.

[0099] In a hardware implementation, the processing units used to execute the techniques may be implemented within one or more ASICs, DSPs, digital signal processing devices (DSPDs), programmable logic devices (PRLs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to perform the functions described in the present disclosure, computers, or combinations thereof.

[0100] Accordingly, the various illustrative logical blocks, modules, and circuits described in connection with the present disclosure may be implemented or executed by any combination of a general purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of devices designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other configuration.

[0101] In a firmware and / or software implementation, the techniques may be realized as instructions stored on a computer-readable medium such as a random access memory (RAM), read-only memory (ROM), nonvolatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable PROM (EEPROM), flash memory, compact disc (CD), magnetic or optical data storage device, etc. The instructions may be executable by one or more processors and may also cause a processor to perform certain aspects of the functions described in the present disclosure.

[0102] When implemented in software, the techniques may be stored on a computer-readable medium as one or more instructions or code, or transferred via a computer-readable medium. A computer-readable medium includes any medium that facilitates transfer of a computer program from one place to another and includes both computer storage media and communication media. A storage media may also be any available media that can be accessed by a computer. By way of non-limiting example, such computer-readable media can include RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage, or other magnetic storage devices, or any other medium that can be used to transfer or store the desired program code in the form of instructions or data structures and that can be accessed by a computer. Further, any connection is properly termed a computer-readable medium.

[0103] For example, if software is transferred from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included within the definition of a medium. As used herein, disk and disc include CD, laser disk, optical disk, DVD (Digital Versatile Disc), floppy disk, and Blu-ray disc where disks typically reproduce data magnetically while discs reproduce data optically using a laser. Combinations of the above should also be included within the scope of computer-readable media.

[0104] The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other known form of storage medium. An exemplary storage medium may be connected to the processor such that the processor can read information from, or write information to, the storage medium. Alternatively, the storage medium may be integrated with the processor. The processor and the storage medium may be present in an ASIC. The ASIC may be present in a user terminal. Or, the processor and the storage medium may be present as separate components in a user terminal.

[0105] The embodiments described above have been described as utilizing aspects of the presently disclosed subject matter in one or more stand-alone computer systems, but the present disclosure is not so limited and may be implemented in conjunction with any computing environment such as a network or a distributed computing environment. Further, aspects of the subject matter in the present disclosure may be implemented in multiple process chips or devices, and storage may be similarly affected across multiple devices. Such devices may also include PCs, network servers, and portable devices.

[0106] Although the present disclosure has been described in connection with some embodiments, various modifications and changes can be made without departing from the scope of the present disclosure that can be understood by those of ordinary skill in the art to which the invention of the present disclosure pertains. Also, such modifications and changes should be considered to fall within the scope of the appended claims of this specification.

Description of Reference Numerals

[0107] 100: User 110: User Terminal 120: Search Result Page 130: Search Term Input Section 140: POI Dataset 142: Review Image Dataset 144: Review Text Dataset

Claims

1. In a method for providing search results that reflects the intentions of a user related to a location, which is executed by at least one processor, receiving a search query from a user terminal; extracting a plurality of POI (Point of Interest) data sets related to the search query; generating a search result page based on the plurality of POI data sets, wherein the plurality of POI data sets related to the search query includes a plurality of review image data sets related to the search query and a plurality of review text data sets related to the search query, the method for providing search results.

2. The step of extracting a plurality of POI data sets related to the search query includes mapping a label to each of a plurality of candidate review image data sets using a first machine learning model; calculating a similarity between the label and the search query; determining at least a part of the plurality of candidate review image data sets as the plurality of POI data sets based on the similarity, The method for providing search results according to claim 1.

3. The first machine learning model is trained to cluster a learning image data set into a plurality of clusters and map a label to each of the plurality of clusters, The step of mapping a label to each of a plurality of candidate review image data sets using the first machine learning model includes calculating a distance value between each of the plurality of candidate review image data sets and each of the plurality of clusters using the first machine learning model; mapping the label of the cluster whose distance value is equal to or less than a threshold value to each of the plurality of candidate review image data sets, The method for providing search results according to claim 2.

4. The label includes a top category label, the method for providing search results according to claim 2.

5. The plurality of candidate review image data sets are generated in relation to a plurality of candidate review text data sets, the method for providing search results according to claim 2.

6. The step of extracting a plurality of POI data sets related to the search query includes mapping a label to each of a plurality of candidate review text data sets using a second machine learning model; calculating a similarity between the label and the search query; Based on the similarity, determining at least a part of the plurality of candidate review text datasets as the plurality of POI datasets; The search result providing method according to claim 1, including this.

7. The step of mapping a label to each of the plurality of candidate review text datasets using the second machine learning model includes: Tokenizing the plurality of candidate review text datasets into a plurality of tokens; Extracting an embedding vector including semantic information from each of the plurality of tokens using the second machine learning model; Mapping a label to each of the plurality of candidate review text datasets based on the embedding vector using the second machine learning model; The search result providing method according to claim 6, including this.

8. The step of generating a search result page based on the plurality of POI datasets includes: Determining a ranking for each of the plurality of POI datasets; Determining a ranking for each of the plurality of POI data included in the plurality of POI datasets; Generating the search result page based on the ranking for each of the plurality of POI datasets and the ranking for each of the plurality of POI data; The search result providing method according to claim 1, including this.

9. The step of determining a ranking for each of the plurality of POI datasets includes: Determining a ranking for each of the plurality of POI datasets based on at least one of the number of POI data, the appearance frequency of keywords, or the appearance rate of keywords The search result providing method according to claim 8, including this.

10. The step of determining the ranking of the plurality of POI data includes: Determining a ranking for each of the plurality of review image data included in the review image dataset related to the search query; Determining a ranking for each of the plurality of review text data included in the review text dataset related to the search query; The search result providing method according to claim 8, including this.

11. The step of determining a ranking for each of the plurality of review text data and the step of determining a ranking for each of the plurality of review image data are executed independently of each other. The search result providing method according to claim 10.

12. The step of determining a ranking for each of the plurality of review image data is The step of determining a ranking for each of the plurality of review image data based on the degree of similarity or recency with the search query The search result providing method according to claim 10, comprising.

13. The step of determining a ranking for each of the plurality of review text data is The step of determining a ranking for each of the plurality of review text data based on at least one of recency, the number of occurrences of keywords, or the amount of information contained in the text data The search result providing method according to claim 10, comprising.

14. The step of generating a search result page based on the plurality of POI data sets is The step of extracting a plurality of snippets from the plurality of POI data sets and The step of generating the search result page so as to include a visual effect of emphasizing at least a part of the plurality of snippets related to the search query The search result providing method according to claim 1, comprising.

15. The search result page includes a keyword filter list for filtering search results, The method is The step of receiving user input for selecting at least one keyword filter from the keyword filter list from the user terminal and The step of regenerating the search result page based on the selected keyword filter in response to the user input The search result providing method according to claim 1, comprising.

16. The step of regenerating the search result page based on the selected keyword filter is The step of determining a ranking for each of the plurality of POI data sets based on the selected keyword filter and The step of regenerating the search result page by sorting the plurality of POI data sets based on the determined ranking The search result providing method according to claim 15, comprising.

17. Each of the plurality of POI data sets is Associated with keyword statistical data including the selection results of keywords related to the specific location by visitors to the specific location The step of determining the ranking of each of the plurality of POI data sets based on the selected keyword filter is as follows: The step of determining the ranking of each of the plurality of POI data sets based on the selected keyword filter and the keyword statistical data The method for providing search results according to claim 16, including this step.

18. The keyword filter directory is as follows: The method for providing search results according to claim 15, including a keyword filter related to the plurality of POI data sets and a personalized keyword filter.

19. A computer-readable computer program for a computer to execute the method according to any one of claims 1 to 18.

20. An information processing system, including: A communication module; A memory; At least one processor configured to execute at least one computer-readable program included in the memory and connected to the memory; Including: The at least one program includes: Receiving a search query from a user terminal; Extracting a plurality of POI data sets related to the search query; Including instructions for generating a search result page based on the plurality of POI data sets; The plurality of POI data sets related to the search query include: An information processing system including a plurality of review image data sets related to the search query and a plurality of review text data sets related to the search query.

Images