Server and method for processing consumer comments

Abstract

Aspects relate to a server for processing consumer reviews, the server configured to: access consumer reviews associated with a service provider; selecting at least one consumer comment related to at least one predetermined category from the consumer comments; obtaining an annotation associated with the selected consumer comment from a computing device associated with the third party; generating tag content associated with the selected consumer comments by summarizing the selected consumer comments; and generating a tag associated with the selected consumer comment based on the tag content and the annotation associated with the selected consumer comment, where the processor is further configured to classify the selected consumer comment based on at least one attribute of the selected consumer comment, and generate a tag associated with the selected consumer comment based on the tag content and the annotation associated with the selected consumer comment. And distributing a task for the annotation associated with the selected consumer comment to a computing device associated with the third party based on the classification of the selected consumer comment.

Description

Technical Field

[0001] Various embodiments are directed to servers and methods for processing consumer reviews. Background Art

[0002] Due to the development of information and communication technology, consumers can use computing devices to request on-demand services. On-demand services can allow consumers to meet their needs via instant access to goods and / or services provided by service providers. Consumers can use user interface screens presented on computing devices to request on-demand services, such as food delivery services.

[0003] Consumers may tend to choose more familiar service providers when requesting on-demand services. Therefore, service providers with a long-tail strategy may face challenges in acquiring new consumers. Due to the low popularity and awareness of service providers with a long-tail strategy, service providers with a long-tail strategy may be less likely to match search keywords entered by consumers and less likely to appear high in the search list of service providers.

[0004] Meanwhile, traditionally, platforms for providing on-demand services may use user-generated content (UGC) previously generated by consumers and associated with service providers (hereinafter, referred to as “consumer reviews”) to enhance the credibility and social recognition of the service providers.

[0005] Figure 1 1 shows exemplary user interface screens 160a, 160b, 160c of a computing device 160 associated with a user 161 (also referred to as a "consumer") in accordance with conventional techniques. Figure 1As shown, the computing device 160 of the user 161 may display a user interface screen 160a for the user 161 to make a request for an on-demand service. The computing device 160 may display a service provider list 191. Information about the service provider may be displayed on a service provider card (also referred to as a "merchant card") 191a, 191b, 191c in the service provider list 191. For example, each service provider card 191a, 191b, 191c may show at least one of promotional information, the type of service provider, the performance of the service provider, an estimated time of arrival ("ETA"), a rating, and information about items provided by the service provider. If the user 161 selects one of the service provider cards 191a, 191b, 191c, such as the first service provider card 191a, the computing device 160 may display a user interface screen 160b showing detailed information about the selected service provider and the items provided by the selected service provider. If the user selects the icon 192 for "View Details," the computing device 160 may display a user interface screen 160c showing consumer reviews 193 associated with the selected service provider.

[0006] However, according to conventional technology, after selecting the first service provider card 191a from the service provider list 191, the user 161 may view only consumer reviews 193 associated with a specific service provider (e.g., the first service provider). If the user 161 is not familiar with the first service provider, the user 161 may be unlikely to select the first service provider card 191a to view consumer reviews 193 associated with the first service provider. In addition, some consumer reviews 193 (e.g., the first consumer review 193a) may be long and require effort from the user 161 to read through. Sometimes, consumer reviews 193 not related to the first service provider may be shown to the user 161, for example, a second consumer review 193b related to a delivery service.

[0007] In addition, when a consumer has a bad experience, the consumer may leave a negative consumer review associated with the service provider, and such negative consumer review may lead to a biased first impression about the service provider. In such a case, the service provider may have to ask the platform to remove the negative consumer review, which may have been displayed and caused some negative impact.

[0008] Therefore, there is a need to provide improved technical solutions for processing consumer reviews. Summary of the invention

[0009] According to various embodiments, there is a server for processing consumer reviews, which includes: a memory for storing instructions; and a processor for executing the stored instructions and is configured to: access consumer reviews associated with a service provider; select at least one consumer review related to at least one predetermined category among the consumer reviews; obtain annotations associated with the selected consumer review from a computing device associated with a third party; generate tag content associated with the selected consumer review by summarizing the selected consumer reviews; and generate tags associated with the selected consumer reviews based on the tag content and annotations associated with the selected consumer reviews, wherein the processor is also configured to classify the selected consumer reviews based on at least one attribute of the selected consumer reviews, and distribute tasks for the annotations associated with the selected consumer reviews to computing devices associated with the third party based on the classification of the selected consumer reviews.

[0010] In some implementations, the processor is further configured to update a rule for selecting at least one of the consumer reviews that is associated with at least one predetermined category based on annotations obtained from a computing device associated with a third party.

[0011] In some implementations, the processor is configured to generate tag content associated with the selected consumer review based on at least one constraint stored in the tag configuration cache.

[0012] In some implementations, the processor is further configured to check whether the tag content satisfies at least one constraint stored in the tag configuration cache, and generate a tag associated with the selected consumer review if the tag content satisfies the at least one constraint.

[0013] In some implementations, the processor is configured to generate tags associated with the selected consumer reviews further based on search keywords entered by the plurality of consumers.

[0014] In some implementations, the processor is further configured to determine that the selected consumer review is related to two or more categories, and to extract two or more phrases from the selected consumer review using a natural language processing model, each phrase being related to the two or more categories.

[0015] In some implementations, the processor is further configured to generate two or more tag contents, each tag content being associated with the two or more phrases.

[0016] In some implementations, the processor is further configured to display tags associated with the selected consumer review and information about the service provider included in the service provider list.

[0017] In some embodiments, the processor is also configured to monitor user behavior with respect to tags displayed on a computing device associated with the user and / or at least one consumer review previously made by the user, and determine which of multiple tags to display on the computing device associated with the user based on the monitored information.

[0018] In some implementations, the processor is further configured to determine a weight for each of the plurality of tags based on the monitored information.

[0019] According to various embodiments, there is a method for processing consumer reviews, the method comprising: accessing consumer reviews associated with a service provider; selecting at least one consumer review from the consumer reviews that is related to at least one predetermined category; classifying the selected consumer reviews based on at least one attribute of the selected consumer reviews; distributing tasks for annotations associated with the selected consumer reviews to a computing device associated with a third party based on the classification of the selected consumer reviews; obtaining annotations associated with the selected consumer reviews from the computing device associated with the third party; generating tag content associated with the selected consumer reviews by summarizing the selected consumer reviews; and generating tags associated with the selected consumer reviews based on the tag content and annotations associated with the selected consumer reviews.

[0020] In some implementations, the method further includes updating a rule for selecting at least one of the consumer reviews that is associated with at least one predetermined category based on annotations obtained from a computing device associated with the third party.

[0021] In some implementations, generating tag content associated with the selected consumer review is based on at least one constraint stored in a tag configuration cache.

[0022] In some implementations, the method further includes: checking whether the tag content satisfies at least one constraint stored in a tag configuration cache; and generating a tag associated with the selected consumer review if the tag content satisfies the at least one constraint.

[0023] In some implementations, generating tags associated with the selected consumer reviews is also based on search keywords entered by the plurality of consumers.

[0024] In some implementations, the method further includes: determining that the selected consumer review is related to two or more categories; and extracting two or more phrases from the selected consumer review using a natural language processing model, each phrase being related to the two or more categories.

[0025] In some implementations, the method further includes: generating two or more tag contents, each tag content being associated with the two or more phrases.

[0026] In some implementations, the method further includes displaying tags associated with the selected consumer review and information about the service provider included in the service provider list.

[0027] In some embodiments, the method further includes: monitoring user behavior with respect to tags displayed on a computing device associated with the user and / or at least one consumer review previously made by the user; and determining which of a plurality of tags to display on the computing device associated with the user based on the monitored information.

[0028] In some implementations, the method further includes determining a weight of each of the plurality of tags based on the monitored information.

[0029] According to various embodiments, there is provided a data processing device configured to execute the method of any one of the above embodiments.

[0030] According to various embodiments, a computer program element is provided, which includes program instructions, which, when executed by one or more processors, cause the one or more processors to perform the method of any one of the above embodiments.

[0031] According to various embodiments, a computer readable medium is provided, which includes program instructions, which when executed by one or more processors cause the one or more processors to perform any of the methods in the above embodiments. The computer readable medium may include a non-transitory computer readable medium. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The invention will be better understood with reference to the detailed description when considered in conjunction with the non-limiting examples and accompanying drawings, in which:

[0033] - Figure 1 Exemplary user interface screens of a computing device associated with a user are shown in accordance with conventional techniques.

[0034] - Figure 2 An infrastructure of a system including a server for processing consumer reviews is shown according to various embodiments.

[0035] - Figure 3 A block diagram of a server for processing consumer reviews is shown in accordance with various embodiments.

[0036] - Figure 4 A flow chart of a method for processing consumer reviews according to various implementations is shown.

[0037] - Figure 5 Exemplary user interface screens of a computing device associated with a user are shown in accordance with various implementations.

[0038] - Figure 6 Exemplary user interface screens of a computing device associated with a user are shown in accordance with various implementations.

[0039] - Figure 7 A block diagram of a system for processing consumer reviews is shown in accordance with various implementations. DETAILED DESCRIPTION

[0040] The following detailed description refers to the accompanying drawings, which illustrate in a diagrammatic manner the specific details and embodiments in which the present disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present disclosure. Other embodiments may also be utilized, and structural and logical changes may be made without departing from the scope of the present disclosure. The various embodiments are not necessarily mutually exclusive, as some embodiments may be combined with one or more other embodiments to form new embodiments.

[0041] Embodiments described in the context of one of the server and the method are also valid for the other of the server and the method. Similarly, embodiments described in the context of the server are also valid for the method, and vice versa.

[0042] Features described in the context of one embodiment may be correspondingly applicable to the same or similar features in other embodiments. Features described in the context of one embodiment may be correspondingly applicable to other embodiments, even if not explicitly described in these other embodiments. In addition, additions and / or combinations and / or alternative means described for features in the context of one embodiment may be correspondingly applicable to the same or similar features in other embodiments.

[0043] In the context of various embodiments, the articles “a,” “an,” and “the” used with respect to features or elements include reference to one or more features or elements.

[0044] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0045] Hereinafter, embodiments will be described in detail.

[0046] Figure 2 The infrastructure of a system 200 including a server 100 for processing consumer reviews according to various embodiments is shown.

[0047] like Figure 2As shown, the system 200 may include, but is not limited to, the server 100, the database system 140, the network 150, and one or more external devices 170 (not shown) associated with one or more service providers 171. In some embodiments, the system 200 may also include a computing device 160 associated with a user 161. In some embodiments, the system 200 may also include one or more computing devices 180 associated with one or more third parties 181.

[0048] In some implementations, an on-demand service may be a service that allows user 161 to satisfy a need of user 161 via instant access to goods and / or services provided by service provider 171. User 161 may request an on-demand service, such as an item delivery service (e.g., a food delivery service), using a user interface screen presented on computing device 160.

[0049] In some embodiments, the network 150 may include, but is not limited to, a local area network (LAN), a wide area network (WAN), a global area network (GAN), or any combination thereof. The network 150 may provide wired communication, wireless communication, or a combination of wired and wireless communication between the server 100 and the computing device 160, between the server 100 and one or more external devices 170 (e.g., one or more service provider devices 170), and between the server 100 and one or more computing devices 180.

[0050] In some embodiments, computing device 160 may be connected to server 100 via network 150. In some embodiments, computing device 160 may be arranged to communicate data or signals with server 100 via network 150. In some embodiments, computing device 160 may include, but is not limited to, at least one of a mobile phone, a tablet computer, a laptop computer, a desktop computer, a head-mounted display, and a smart watch. In some embodiments, computing device 160 may be associated with user 161. For example, computing device 160 may belong to user 161. Although not shown, in some embodiments, system 200 may also include multiple computing devices, each of which belongs to multiple users. In some embodiments, user 161 may be a consumer. For example, user 161 may leave a consumer review associated with a service provider on a platform operated by server 100 and providing on-demand services.

[0051] In some embodiments, one or more computing devices 180 may be capable of connecting to the server 100 via the network 150. In some embodiments, the one or more computing devices 180 may be arranged to communicate data or signals with the server 100 via the network 150. In some embodiments, the one or more computing devices 180 may include, but are not limited to, at least one of the following: a mobile phone, a tablet computer, a laptop computer, a desktop computer, a head-mounted display, and a smart watch. In some embodiments, the one or more computing devices 180 may be associated with one or more third parties 181, respectively. For example, each computing device in the one or more computing devices 180 may belong to each of the one or more third parties 181, respectively. Although not shown, in some embodiments, the one or more third parties 181 may be consumers. In some other embodiments, one or more third parties 181 may not be consumers. In some other embodiments, a portion of the one or more third parties 181 may be a consumer, and other portions of the one or more third parties 181 may not be consumers. In some embodiments, the one or more third parties 181 may use a predetermined software application installed in the one or more computing devices 180, such as a messaging program (e.g., a Slack application).

[0052] In some embodiments, for example, the server 100 implemented by a server computer may include a communication interface 110, a processor 120, and a memory 130 (as will be referred to in Figure 3 described).

[0053] In some embodiments, the system 200 may also include a database 141. In some embodiments, the database 141 may be part of a database system 140 that may be external to the server 100. The server 100 may communicate with the database 141. In some other embodiments, although not shown, the database 141 may be implemented locally in the memory 130 of the server 100.

[0054] In some embodiments, consumers who use on-demand services can leave consumer reviews associated with service providers 171 on the platform that provides on-demand services. For example, consumers (which may include user 161) can leave consumer reviews associated with first service provider 171a after ordering food and receiving food from first service provider 171a. As an example, consumers can leave consumer reviews related to first service provider 171a, such as the price, taste, safety, packaging and / or service of first service provider 171a. As another example, consumers can leave consumer reviews that are not related to first service provider 171a. For example, consumer reviews can be related to delivery services for delivering food provided by first service provider 171a. As another example, consumers can leave consumer reviews that include content related to first service provider 171a and content that is not related to first service provider 171a. Consumers can type consumer reviews using his / her computing device. In some embodiments, the server 100 of the operating platform can encourage consumers who order food and receive food to leave consumer reviews. For example, once delivery of food provided by the first service provider 171a is completed, the server 100 of the operating platform may control the consumer's computing device to display a pop-up window so that the consumer may easily leave a consumer review associated with the first service provider 171a.

[0055] In some embodiments, the consumer reviews previously left by the consumer may be stored in the memory 130 of the server 100 and / or in the database 141 of the database system 140. In some other embodiments, the consumer reviews may be stored in an external database (not shown), and the communication interface 110 of the server 100 may access the external database.

[0056] In some embodiments, the server 100 can access consumer reviews associated with the service provider 171 and process the consumer reviews (e.g., Figure 3 described).

[0057] Figure 3 A block diagram of a server 100 for processing consumer reviews according to various embodiments is shown.

[0058] like Figure 3 As shown, the server 100 , which is implemented by a server computer, for example, may include a communication interface 110 , a processor 120 , and a memory 130 .

[0059] In some embodiments, the memory 130 (also referred to as a "database") can temporarily or permanently store input data and / or output data. In some embodiments, the memory 130 can store data that allows the server 100 to perform the method 300 (as described in reference to Figure 4 In some embodiments, the program code may be embedded in a software development kit (SDK). The memory 130 may include an internal memory and / or an external memory of the server 100. The external memory may include, but is not limited to, an external storage medium such as a memory card, a flash drive, and a network memory.

[0060] In some embodiments, the communication interface 110 may allow one or more computing devices (including computing device 160 and one or more computing devices 180) to communicate with the processor 120 of the server 100 via the network 150, such as Figure 2 In some embodiments, Figure 2 As shown, computing device 160 may belong to user 161 who wants to request an on-demand service, and one or more computing devices 180 may belong to one or more third parties 181 who provide annotations (also referred to as "markups") to consumer reviews. In some implementations, communication interface 110 may send signals to and / or receive signals from computing device 160 and one or more computing devices 180 via network 150.

[0061] In some embodiments, the communication interface 110 may allow one or more external devices 170 (eg, one or more service provider devices 170 ) to communicate with the processor 120 of the server 100 via the network 150 , such as Figure 2 In some implementations, the communication interface 110 may send signals to and / or receive signals from one or more external devices 170 via the network 150 .

[0062] In some embodiments, the processor 120 may include, but is not limited to, a microprocessor, an analog circuit, a digital circuit, a mixed signal circuit, a logic circuit, an integrated circuit, a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or any combination thereof. Any other type of implementation of the corresponding functions described in further detail below may also be understood as the processor 120.

[0063] In some embodiments, the processor 120 may access consumer reviews associated with a service provider 171 (e.g., a first service provider 171a). In some embodiments, the processor 120 may access the memory 130 of the server 100 and / or the database 141 of the database system 140 via the communication interface 110 to access the consumer reviews. For example, the processor 120 may obtain consumer reviews stored in the memory 130 of the server 100 and / or the database 141 of the database system 140 via the communication interface 110. In some other embodiments, the processor 120 may access an external database storing consumer reviews via the communication interface 110. For example, the processor 120 may obtain consumer reviews stored in an external database via the communication interface 110. In some embodiments, the processor 120 may access at least a portion of the consumer reviews stored in the memory 130 of the server 100, the database 141 of the database system 140 via the communication interface 110, and / or access at least a portion of the consumer reviews stored in the external database via the communication interface 110. As an example, the processor 120 may decide which consumer reviews the processor 120 will access and / or obtain, for example, based on the date the consumer review was left.

[0064] In some embodiments, after accessing the consumer reviews, the processor 120 may select at least one of the consumer reviews that is associated with at least one predetermined category. In some embodiments, the processor 120 may include a natural language processing engine that is configured to analyze the consumer reviews and select at least one of the consumer reviews based on the relevance to at least one predetermined category. As described above, in some embodiments, the consumer reviews associated with the first service provider 171a may be associated with the first service provider 171a, for example, the price, taste, safety, packaging and / or service of the first service provider 171a. In some other embodiments, the consumer reviews associated with the first service provider 171a may be partially associated with the first service provider 171a. In some other embodiments, the consumer reviews associated with the first service provider 171a may not be associated with the first service provider 171a.

[0065] In some embodiments, the processor 120 may predetermine at least one category associated with the service provider. For example, the predetermined category may include, but is not limited to, the price, taste, safety, packaging, and / or service of the service provider. It is understood that in some embodiments, the predetermined category may be updated based on the processor 120 and / or the input from the operator of the platform. In some embodiments, the processor 120 may use a natural language processing engine to analyze consumer comments, determine whether each consumer comment in the consumer comments is related to at least one predetermined category, and select at least one consumer comment in the consumer comments that is related to at least one predetermined category. For example, if the first consumer comment states "I am very satisfied with the taste of the food", the processor 120 may select the first consumer comment related to the "taste" category. As another example, if the second consumer comment states "the delivery driver is very friendly", the processor 120 may not select the second consumer comment that is not related to any of the predetermined categories in the predetermined categories. As another example, if the third consumer comment states "I am very satisfied with the taste of the food and the friendliness of the delivery driver", the processor 120 may select the third consumer comment related to the "taste" category. As another example, if the fourth consumer review states "I am very satisfied with the taste and reasonable price of the food", the processor 120 may select the fourth consumer review related to the "taste" category and the "price" category. In this way, consumer reviews that are not related to the service provider may be filtered out.

[0066] In some embodiments, the processor 120 can classify the selected consumer review based on at least one attribute of the selected consumer review, and distribute tasks for annotations associated with the selected consumer review to at least one computing device among one or more computing devices 180 associated with one or more third parties 181 based on the classification of the selected consumer review.

[0067] In some embodiments, one or more third parties 181 may use a predetermined software application installed in one or more computing devices 180, such as a messaging program (e.g., a Slack application). A server (not shown) operating the predetermined software application may receive a task for an annotation from the processor 120, receive the annotation from the one or more third parties 181, and send the annotation to the processor 120.

[0068] In some embodiments, the processor 120 may classify the selected consumer reviews based on at least one attribute of the metadata of the selected consumer reviews. In some embodiments, the processor 120 may apply high-level classification on at least one attribute to the selected consumer reviews. In some embodiments, at least one attribute may include, but is not limited to, business verticals, topics, and languages. For example, if the first consumer review states in English, "I am very satisfied with the taste of the food," and the first consumer review is associated with the first service provider 171a (e.g., A donut shop), the processor 120 may classify the first consumer review as a business vertical of "dessert bakery," a topic of "taste," and a language of "English." As another example, if the fifth consumer review states “Semoga murah rezeki (May good fortune and prosperity)” in Bahasa (e.g., due to the word “livelihood (rezeki)”, the actual meaning of the sentence is not related to price but rather to blessings), and the fifth consumer review is associated with the first service provider 171a (e.g., A donut shop), the processor 120 may classify the fifth consumer review as the business vertical of “dessert bakery”, the subject of “irrelevant” (because the fifth consumer review is related to blessings), and the language of “Bahasa”.

[0069] In some embodiments, the processor 120 can distribute the task of annotating the selected consumer review associated with the selected consumer review to at least one of the one or more computing devices 180 associated with one or more third parties 181 based on the classification of the selected consumer review. In some embodiments, the third party 181 (e.g., an agent and / or an internal employee) can register the annotation program via an internal device application and subscribe to the category and language of the consumer review for which they are happy to give accurate annotations. The processor 120 can push the consumer review based on the frequency of the third party 181 request, and the third party 181 can enter their answer (e.g., annotation) on the consumer review via the device application. The processor 120 can use the answer of the third party 181 for modeling. In this way, the third party associated with the classification can be assigned the task for annotation. For example, if the first third party 181a is related to the business vertical field of "dessert bakery", the theme of "taste", and the language of "English", the first third party 181a can be assigned the task of annotating the first consumer review. As another example, if the second third party 181b is associated with the business vertical of "dessert bakery", the subject of "price", and the language of "Bahasa", the second third party 181b may be assigned the task of annotating the fifth consumer review. Although not shown, in some embodiments, multiple third parties associated with a category may be assigned the task of annotating the same consumer review.

[0070] In some embodiments, the processor 120 may communicate with a server operating a predetermined software application via the communication interface 110. In some embodiments, the processor 120 may send a request to the server operating the predetermined software application for distributing a task for annotating the selected consumer review, and the server operating the predetermined software application may distribute the task for annotating to at least one computing device of one or more computing devices 180 associated with one or more third parties 181 based on the classification of the selected consumer review. For example, the processor 120 may send a request to the server operating the predetermined software application for distributing a task for annotating a first consumer review and information about the classification of the first consumer review. Then, the server operating the predetermined software application may select a third party, such as a first third party 181a, which will perform the task of annotating the first consumer review based on the classification of the first consumer review and the relevance of each third party to the classification, and distribute the task of annotating the first consumer review to a first computing device 180a associated with the first third party 181a.

[0071] In some other embodiments, the processor 120 may select a third party that will perform the task of annotating the selected consumer review and send a request to a server operating a predetermined software application for distributing the task of annotating the selected consumer review to the selected third party. Based on the request of the processor 120, the server operating the predetermined software application may distribute the task of annotating to the selected third party. For example, the processor 120 may select a first third party 181a that will perform the task of annotating the first consumer review and send a request to the server operating the predetermined software application for distributing the task of annotating the first consumer review to the first third party 181a. The server operating the predetermined software application may distribute the task of annotating to the first third party 181a.

[0072] In some embodiments, the processor 120 can obtain annotations associated with the selected consumer review from one or more computing devices 180 associated with one or more third parties 181. In some embodiments, the server operating the predetermined software application can receive annotations from the assigned third party. For example, the server operating the predetermined software application can receive annotations of the first consumer review from the computing device 180a associated with the first third party 181a. In some embodiments, at least one of the language of the consumer review, the category of the consumer review, and one or more short phrases summarized from the consumer review can be an annotation that can be used as a recommendation tag for the service provider. For example, in the case where the language detector is wrong, annotations about language may be needed. For example, the sixth consumer review states "This is the second time I've ordered from here. Love the oat milk latte, and the cake tastes great too! Looking forward to more vegetarian options and more offers." As an example, the annotation of the sixth consumer review can be as follows:

[0073] • Language: English

[0074] • Category: Taste

[0075] • Short phrases: 1. Love the oat milk latte. 2. The delicious cake or cupcake tastes delicious.

[0076] As described above, the processor 120 can schedule tasks for annotations via a predetermined software application based on, for example, business verticals, topics, and languages. In addition, the processor 120 can use a multilingual corpus (e.g., a Southeast Asian multilingual corpus) that can contextualize the meaning of tokens in sentences included in consumer reviews, which may be difficult to accurately translate or learn a robust contextual relationship with other tokens in the same sentence, especially when the sentence is short. For example, if the fifth consumer review states "May wealth and fortune be prosperous (Semoga murahrezeki)" in Bahasa, then since the feedback data store 401 (such as the reference Figure 7 120 ). In the example of FIG. 120 , the processor 120 may misinterpret “rezeki” (e.g., meaning “livelihood”) as “price” with a negative sentiment, and the fifth consumer review may be wrongly determined to be related to the “price” category. However, in reality, the fifth consumer review means a kind of “blessing”. The multilingual corpus can detect that the fifth consumer review means a “blessing” that is not related to the “price” category, and further improve the accuracy of the processor 120.

[0077] In some embodiments, the processor 120 may generate tag content associated with the selected consumer review by summarizing the selected consumer review. In some embodiments, the processor 120 may include a natural language generation engine configured to summarize the content of the selected consumer review and generate the tag content. In some embodiments, the processor 120 may include a tag configuration cache 407 (such as referring to Figure 7 As described above, the label configuration cache is configured to store at least one constraint (also referred to as a "configuration") for generating label content, which may be suitable for a label to be displayed on a service provider card. For example, at least one constraint may include, but is not limited to, content freshness, maximum number of words or length, and language. In some embodiments, the processor 120 may generate label content associated with the selected consumer review based on at least one constraint stored in the label configuration cache 407. For example, if the first consumer review states "I am very satisfied with the taste of the food", the processor 120 may generate label content of "good taste" by summarizing the content of the first consumer review of "I am very satisfied with the taste of the food". As an example, the processor 120 may check the constraints for generating label content, such as whether the language is English and whether the maximum number of words of the label content does not exceed a predetermined number. If the generated label content does not meet the constraints, the processor 120 may modify the generated label content to meet the constraints.

[0078] In some embodiments, the processor 120 may generate a tag associated with the selected consumer review based on the tag content and annotations associated with the selected consumer review. In some embodiments, the processor 120 may check whether the tag content satisfies the constraints stored in the tag configuration cache 407, such as content freshness, maximum word count or length, and language, and if the tag content satisfies at least one constraint, a tag associated with the selected consumer review is generated. In some embodiments, the processor 120 may also generate a tag associated with the selected consumer review based on search keywords entered by multiple consumers. For example, an RNN (recurrent neural network) model may be used to integrate information of search keywords entered by multiple consumers to generate a tag or prompt associated with a service provider that is most relevant to the search keyword.

[0079] In some implementations, the processor 120 may display the generated tag associated with the selected consumer review and information about the service provider included in the service provider list. For example, the processor 120 may generate a tag based on the tag content of "good taste" and the received comment for display on the computing device 160 of the user 161 who sets the language of the platform to English and is searching for dessert. The processor 120 may then display the generated "good taste" tag on the first service provider card in the service provider list.

[0080] In some embodiments, the selected consumer comments may be associated with two or more categories. The processor 120 may determine that the selected consumer comments are associated with two or more predetermined categories. For example, if the fourth consumer comment states "I am very satisfied with the taste and reasonable price of the food", the processor 120 may select a fourth consumer comment associated with the "taste" category and the "price" category. In some embodiments, the processor 120 may extract two or more phrases from the selected consumer comments using a natural language processing model, each phrase being associated with the two or more categories. For example, the processor 120 may extract a phrase "I am very satisfied with the taste of the food" associated with the "taste" category and a phrase "I am very satisfied with the reasonable price" associated with the "price" category from the fourth consumer comment. As an example, the processor 120 may use a word segmenter (e.g., a BERT word segmenter). In some embodiments, the processor 120 may generate two or more tag contents, each of which is associated with the two or more phrases. For example, the processor 120 may generate tag contents of "good taste" and "reasonable price" according to the extracted phrases, respectively. In some implementations, the processor 120 may generate a tag based on the two tag contents and the received annotation.

[0081] In some embodiments, the processor 120 may update a rule based on annotations obtained from one or more computing devices 180 associated with one or more third parties 181, the rule being used to select at least one consumer review associated with at least one predetermined category in the consumer reviews. In some embodiments, the processor 120 may select at least one consumer review associated with at least one predetermined category in the consumer reviews based on a rule (also referred to as a "predetermined rule"). After receiving the annotation from a server operating a predetermined software application, the processor 120 may update the predetermined rule based on the annotation to improve the predetermined rule. In other words, the annotation may be used as an input to the training processor 120 to update the predetermined rule.

[0082] In some implementations, processor 120 may monitor user 161's behavior with respect to tags displayed on computing device 160 and / or at least one consumer review previously made by user 161. Processor 120 may determine, based on the monitored information, which of the multiple tags to display on computing device 160. For example, if processor 120 may determine that user 161 is interested in price rather than taste, processor 120 may display price-related tags on computing device 160.

[0083] In some embodiments, the processor 120 may determine a weight for each of the plurality of tags based on the monitored information. For example, if the processor 120 may determine that the user 161 is interested in price rather than taste, the processor 120 may assign a higher weight to tags related to price and assign a lower weight to tags related to other categories (e.g., taste, service, etc.).

[0084] Figure 4 A flow chart of a method 300 for processing consumer reviews is shown, according to various implementations.

[0085] According to various implementations, a method 300 for processing consumer reviews may be provided.

[0086] In some implementations, method 300 may include step 301: accessing consumer reviews associated with a service provider.

[0087] In some implementations, method 300 may include step 302: selecting at least one consumer review among the consumer reviews that is related to at least one predetermined category.

[0088] In some implementations, method 300 may include step 303 : categorizing the selected consumer review based on at least one attribute of the selected consumer review.

[0089] In some implementations, method 300 may include step 304 of distributing a task for an annotation associated with the selected consumer review to a computing device associated with a third party based on the classification of the selected consumer review.

[0090] In some implementations, method 300 may include step 305 of obtaining annotations associated with the selected consumer review from a computing device associated with a third party.

[0091] In some implementations, method 300 may include step 306 of generating tag content associated with the selected consumer review by summarizing the selected consumer review.

[0092] In some implementations, method 300 may include step 307 of generating a tag associated with the selected consumer review based on the tag content and annotations associated with the selected consumer review.

[0093] Figure 5 An exemplary user interface screen 160d of a computing device 160 associated with a user 161 is shown in accordance with various implementations.

[0094] like Figure 5 As shown, a computing device 160 associated with a user 161 may display a user interface screen 160d for the user 161 to make a request for an on-demand service. The user interface screen 160d may show a service provider list 191. Information about the service provider may be displayed on service provider cards 191a, 191b, 191c in the service provider list 191. For example, each service provider card 191a, 191b, 191c may show at least one of promotional information, a type of service provider, performance of the service provider, an estimated time of arrival (“ETA”), a rating, and information about items provided by the service provider.

[0095] In addition, if Figure 5 As shown, each service provider card 191a, 191b, 191c may show a label 194 associated with the corresponding service provider. For example, a first service provider card 191a associated with a first service provider (e.g., A donut shop) may show a label 194a generated based on a consumer review of the first service provider. As an example, a second service provider card 191b associated with a second service provider (e.g., B coffee and tea) may show a label 194b generated based on a consumer review of the second service provider. As an example, a third service provider card 191c associated with a third service provider (e.g., C bakery) may show a label 194c generated based on a consumer review of the third service provider.

[0096] Figure 6 An exemplary user interface screen 160e of a computing device 160 associated with a user 161 is shown in accordance with various implementations.

[0097] In some embodiments, the processor 120 of the server 100 may monitor the behavior of the user 161 with respect to the tags displayed on the computing device 160 and / or at least one consumer review previously made by the user 161. The processor 120 may determine which of the multiple tags to display on the computing device 160 based on the monitored information (e.g., reference to the tag). Figure 3 described).

[0098] In this way, if Figure 6As shown, the tags 194 displayed on the service provider cards 191a, 191b, 191c may be related to the interests of the user 161. For example, if the processor 120 determines based on the monitored information that the user 161 is interested in healthy food, the processor 120 may select a tag related to "healthy food" and display the selected tag on the service provider cards 191a, 191b, 191c.

[0099] Figure 7 A block diagram of a system 200 for processing consumer reviews is shown, according to various implementations.

[0100] like Figure 7 As shown, the system 200 may include a tag content providing unit 400a, a tag management providing unit 400b, and a front end 400c.

[0101] In some embodiments, the feedback data store (VoC (Voice of the Consumer)) 401 may be a raw data source for consumer reviews. In some embodiments, the feedback data store 401 may store metadata for consumer reviews and apply high-level classifications with respect to business verticals, topics, and / or language. In some embodiments, the feedback data store 401 may store data on a daily basis.

[0102] In some embodiments, the data annotation outsourcing system 402 may use a predetermined software application (e.g., a Slack application) 403. In some embodiments, a server operating the predetermined software application 403 may be connected to a server operating a predetermined network application (e.g., an Azure network application) and a predetermined database (e.g., an SQL database), which may refresh and query data based on filters and / or rules regarding business verticals, topics, and / or languages ​​received from the feedback data storage 401. In some embodiments, the server operating the predetermined software application 403 may communicate directly with a third party (e.g., a user of the Slack application) to distribute data annotation tasks for consumer reviews (also referred to as "tasks for annotations"), and receive results (also referred to as "responses") (e.g., including annotations) from the third party. In some embodiments, a server operating a predetermined network application may process requests for data annotation tasks and results received from a third party. In some embodiments, a predetermined database may store results received from a third party. In some embodiments, the predetermined database may also store subscription data of a third party, including but not limited to country, native language, and contribution frequency.

[0103] In some embodiments, the relevance classification engine 404 can be a service driven by a natural language processing model. In some embodiments, the relevance classification engine 404 can accurately identify whether the consumer review is related to the service provider, such as the price, taste, safety, packaging and / or service of the service provider, and send the relevant consumer reviews to the downstream service. In some embodiments, the training data set of the natural language processing model can come from the data annotation outsourcing system 402. In some embodiments, the prediction results of the service can be fed back to the feedback data storage 401, so that the feedback data storage 401 can more efficiently select the original data for the data annotation task based on the improved data filtering rules. In some embodiments, the data annotation outsourcing system 402 can read the original data selected from the feedback data storage 401.

[0104] In some embodiments, the summary content engine 405 can be a service driven by a natural language generation model that can consume relevant consumer reviews received from the relevance classification engine 404 and specific constraints on the result content received from the tag configuration cache 407. In some embodiments, the summary content engine 405 can generate summary content (also called "tag content") and send the summary content to a data source 406 that can be suitable for a tag on a service provider card.

[0105] In some embodiments, the tag management providing unit 400b may perform a "tag creation" step, a "tag management" step, and a "tag display" step. In some embodiments, the tag management providing unit 400b may create and update new tags with summary tag content, thereby allowing configuration flexibility and manual intervention for factors such as content freshness, maximum word count or length, and language.

[0106] In some embodiments, in the "label creation" step, the delv feedback system (delivery-feedback system) 410 can be used. In some embodiments, the delv feedback system 410 can be configured to centralize multiple label management requests, including creation, update, deletion, storage, and rule-based validation. In some embodiments, the delv feedback system 410 can consume label content and related settings received from data sources 406 and other platforms (e.g., GrabX and SegP), and perform a secondary check whether the label content meets constraints (also known as "configuration"), including but not limited to content freshness, maximum word count or length, and language. In some embodiments, if the service provider can enable the service provider label feature based on UGC (based on user-generated content), the delv feedback system 410 can create labels for the service provider and prepare for downstream services.

[0107] In some embodiments, in the "Tag Management" step, content settings for tags of a particular location (e.g., a particular city) or service provider may be defined. For example, the content settings may be defined by an operator of a platform operating the server 100. In some embodiments, configuration flexibility may be allowed regarding content freshness period, maximum word count or length, and language. In some embodiments, rules may be stored in the tag configuration cache 407 for use with the tag content provider 400a and the tag management provider 400b.

[0108] In some embodiments, during the "Tag Display" step, the delv feedback system 410 can send tags to the front end 400c. In some embodiments, the service provider can approve or reject their tags before display, and the service provider's feedback can be recorded for engine optimization. In some embodiments, the notification frequency and batch size of tags for verification can be configurable. In some embodiments, an operator of the platform operating the server 100 can review the tag log and manually modify the tags (e.g., via Zeus). Thereafter, consumers can view tags generated based on consumer reviews, such as Figure 5 and Figure 6 shown.

[0109] In some embodiments, an automated analysis engine (not shown) can provide prior information about the consumer's preferences and transaction behavior. For example, a consumer detected as a "value seeker" may be more likely to see tags under the "price" category. In some embodiments, the automated analysis engine can evaluate the before-and-after performance to make decisions, such as removing tags that may result in low or even negative impacts, for example, via a Bayesian structural time series model (BSTS) and significance testing. For example, the automated analysis engine can evaluate the financial impact after the tag is released to decide on personalized actions for the displayed tags. In some embodiments, the automated analysis engine can fine-tune the weight of each tag and / or shape the consumer's needs.

[0110] As described above, according to various embodiments, the system 200 including the server 100 can help promote the development of service providers with a long-tail strategy and gain consumers' interest and trust in unfamiliar service providers by:

[0111] An annotation outsourcing system and a multilingual corpus (eg, a Southeast Asian multilingual corpus) collected from a chatbot of a predetermined software application (eg, a Slack chatbot) and feedback associated with a service provider triggered by the predetermined software application.

[0112] Filter and summarize relevant consumer reviews through multiple stages of text data processing:

[0113] - The data product (Voice of the Consumer) implemented in the server 100 may run a first round of filtering and embedding clustering to select consumer reviews related to, for example, price, taste, safety, packaging and / or service of a service provider.

[0114] - Natural language processing engines (e.g., leveraging distilBERT and GPT-2 models) can run classification to further filter consumer reviews for outsourcing the annotation task and generate multiple concise labels reviewing different aspects of the service provider.

[0115] An automated merchant tag management system for creating and updating new tags with summary tag content, allowing for configuration flexibility and manual intervention for things like content freshness period, maximum word count or length, and language.

[0116] Although the present disclosure has been specifically shown and described with reference to specific embodiments, it should be understood by those skilled in the art that various changes in form and details may be made to the present disclosure without departing from the spirit and scope of the present invention as defined in the appended claims. Therefore, the scope of the present invention is indicated by the appended claims, and therefore, it is intended to include all changes within the meaning and equivalent range of the claims.

Claims

1. A server for processing consumer reviews, the server comprising: A memory for storing instructions; as well as A processor for executing the stored instructions and configured to: accessing said consumer reviews associated with the service provider; selecting at least one consumer review among the consumer reviews that is associated with at least one predetermined category; obtaining, from a computing device associated with a third party, annotations associated with the selected consumer review; generating tag content associated with the selected consumer review by summarizing the selected consumer review; and generating a tag associated with the selected consumer review based on the tag content and the annotation associated with the selected consumer review, The processor is further configured to classify the selected consumer review based on at least one attribute of the selected consumer review, and to distribute tasks for the annotations associated with the selected consumer review to the computing device associated with the third party based on the classification of the selected consumer review.

2. The server according to claim 1, wherein: The processor is further configured to update a rule for selecting the at least one of the consumer reviews that is associated with the at least one predetermined category based on the annotation obtained from the computing device associated with the third party.

3. The server according to claim 1 or claim 2, wherein: The processor is configured to generate the tag content associated with the selected consumer review based on at least one constraint stored in a tag configuration cache.

4. The server according to claim 3, wherein: The processor is further configured to check whether the tag content satisfies the at least one constraint stored in the tag configuration cache, and to generate the tag associated with the selected consumer review if the tag content satisfies the at least one constraint.

5. The server according to claim 3 or claim 4, wherein: The processor is configured to generate the tags associated with the selected consumer reviews further based on search keywords entered by a plurality of consumers.

6. The server according to any one of claims 1 to 5, wherein: The processor is also configured to determine that the selected consumer review is associated with two or more categories, and to extract two or more phrases from the selected consumer review using a natural language processing model, each phrase being associated with the two or more categories.

7. The server according to claim 6, wherein: The processor is further configured to generate two or more tag contents, each tag content being associated with the two or more phrases.

8. The server according to any one of claims 1 to 7, wherein: The processor is further configured to display the tag associated with the selected consumer review and information about the service provider included in the service provider list.

9. The server according to any one of claims 1 to 8, wherein: The processor is also configured to monitor user behavior with respect to tags displayed on a computing device associated with the user and / or at least one consumer review previously made by the user, and determine which of a plurality of tags to display on the computing device associated with the user based on the monitored information.

10. The server according to claim 9, wherein: The processor is further configured to determine a weight of each of the plurality of tags based on the monitored information.

11. A method for processing consumer reviews, the method comprising: accessing said consumer reviews associated with the service provider; selecting at least one consumer review among the consumer reviews that is associated with at least one predetermined category; categorizing the selected consumer review based on at least one attribute of the selected consumer review; distributing tasks for annotations associated with the selected consumer reviews to a computing device associated with a third party based on the classification of the selected consumer reviews; obtaining, from the computing device associated with the third party, the annotation associated with the selected consumer review; generating tag content associated with the selected consumer review by summarizing the selected consumer review; and A tag associated with the selected consumer review is generated based on the tag content and the annotation associated with the selected consumer review.

12. The method according to claim 11, further comprising: A rule is updated based on the annotation obtained from the computing device associated with the third party, the rule being used to select the at least one of the consumer reviews that is associated with the at least one predetermined category.

13. The method according to claim 11 or claim 12, wherein: Generating the tag content associated with the selected consumer review is based on at least one constraint stored in a tag configuration cache.

14. The method according to claim 13, further comprising: checking whether the tag content satisfies the at least one constraint stored in the tag configuration cache; as well as If the tag content satisfies the at least one constraint, the tag is generated in association with the selected consumer review.

15. The method according to claim 13 or claim 14, wherein: Generating the tags associated with the selected consumer reviews is also based on search keywords entered by the plurality of consumers.

16. The method according to any one of claims 11 to 15, further comprising: Determining that the selected consumer reviews are relevant to two or more categories; as well as Two or more phrases are extracted from the selected consumer reviews using a natural language processing model, each phrase being associated with the two or more categories.

17. The method according to claim 16, further comprising: Two or more tag contents are generated, each tag content being associated with the two or more phrases.

18. The method according to any one of claims 11 to 17, further comprising: The tag associated with the selected consumer review and information about the service provider included in the service provider list are displayed.

19. The method according to any one of claims 11 to 18, further comprising: monitoring a user's behavior with respect to a tag displayed on a computing device associated with the user and / or at least one consumer review previously made by the user; as well as Determining which of a plurality of tabs to display on the computing device associated with the user is based on the monitored information.

20. The method according to claim 19, further comprising: A weight of each of the plurality of tags is determined based on the monitored information.

Images