WO2020177178A1 - 一种智能识别系统中的资源检索方法及智能识别系统 - Google Patents

一种智能识别系统中的资源检索方法及智能识别系统 Download PDF

Info

Publication number
WO2020177178A1
WO2020177178A1 PCT/CN2019/081535 CN2019081535W WO2020177178A1 WO 2020177178 A1 WO2020177178 A1 WO 2020177178A1 CN 2019081535 W CN2019081535 W CN 2019081535W WO 2020177178 A1 WO2020177178 A1 WO 2020177178A1
Authority
WO
WIPO (PCT)
Prior art keywords
resource
retrieval
search
user
identification system
Prior art date
Application number
PCT/CN2019/081535
Other languages
English (en)
French (fr)
Inventor
吴文强
Original Assignee
网宿科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 网宿科技股份有限公司 filed Critical 网宿科技股份有限公司
Priority to EP19820654.2A priority Critical patent/EP3734473A4/en
Priority to US16/726,063 priority patent/US11122308B2/en
Publication of WO2020177178A1 publication Critical patent/WO2020177178A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/25—Integrating or interfacing systems involving database management systems
    • G06F16/256—Integrating or interfacing systems involving database management systems in federated or virtual databases
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G06F16/9032—Query formulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/08—Payment architectures
    • G06Q20/12—Payment architectures specially adapted for electronic shopping systems
    • G06Q20/127—Shopping or accessing services according to a time-limitation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/08—Payment architectures
    • G06Q20/14—Payment architectures specially adapted for billing systems
    • G06Q20/145—Payments according to the detected use or quantity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/232—Content retrieval operation locally within server, e.g. reading video streams from disk arrays
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/41—Structure of client; Structure of client peripherals
    • H04N21/418—External card to be used in combination with the client device, e.g. for conditional access
    • H04N21/4182—External card to be used in combination with the client device, e.g. for conditional access for identification purposes, e.g. storing user identification data, preferences, personal settings or data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/438—Interfacing the downstream path of the transmission network originating from a server, e.g. retrieving encoded video stream packets from an IP network

Definitions

  • This application relates to the field of Internet technology, in particular to a resource retrieval method and an intelligent identification system in an intelligent identification system.
  • the purpose of this application is to provide a resource retrieval method and an intelligent identification system in an intelligent identification system, which can simplify the way for users to obtain resources.
  • this application provides a resource retrieval method in an intelligent identification system
  • the intelligent identification system is associated with multiple resource servers provided by multiple resource service providers, and the method includes: receiving user input Retrieve information, and identify the resource type represented by the retrieval information; determine several target resource servers corresponding to the resource type from among the multiple resource servers, and send the retrieval information to the several target resource servers At; receiving the retrieval resources fed back by the several target resource servers for the retrieval information, and showing the received retrieval resources to the user.
  • another aspect of the present application also provides an intelligent identification system
  • the intelligent identification system is associated with multiple resource servers provided by multiple resource service providers, and the intelligent identification system includes: a retrieval information entry unit, It is used to receive the search information input by the user and identify the resource type represented by the search information; the resource search unit is used to determine a number of target resource servers corresponding to the resource type among the multiple resource servers, and to The retrieval information is sent to the plurality of target resource servers; the resource display unit is configured to receive retrieval resources fed back by the plurality of target resource servers for the retrieval information, and display the received retrieval to the user Resources.
  • the technical solution provided by this application can associate multiple resource servers provided by different resource service providers to the same intelligent identification system, so that the intelligent identification system can provide users with a unified search portal.
  • the intelligent recognition system can preferentially identify the resource type represented by the search information. For example, the smart recognition system can recognize whether the user wants to search for music, videos, or articles.
  • the intelligent identification system can deliver the retrieval information to the corresponding target resource server for processing. Specifically, assuming that the intelligent recognition system recognizes that the user currently wants to retrieve video resources, the retrieval information can be sent to the resource servers of service providers such as Youku Video, Tencent Video, and iQiyi Video for processing.
  • each target resource server After each target resource server processes and obtains the retrieval resources, it can feed the retrieval resources back to the intelligent identification system. In this way, the intelligent identification system can display these retrieval resources to the user for selection. It can be seen from the above that this application provides a unified intelligent identification system. Users do not need to download multiple different resource software to obtain the retrieval resources they are interested in, thereby reducing the user’s use cost and simplifying the user’s retrieval method. .
  • Figure 1 is a schematic diagram of the steps of a resource retrieval method in an embodiment of the present application
  • FIG. 2 is a schematic diagram of the network architecture of the intelligent identification system in the embodiment of the present application.
  • FIG. 3 is a schematic diagram of functional modules of the intelligent identification system in an embodiment of the present application.
  • Fig. 4 is a schematic diagram of internal components of the intelligent identification system in an embodiment of the present application.
  • the present application provides a resource retrieval method in an intelligent identification system, which can interface with multiple resource servers provided by different resource service providers.
  • the resource servers provided by these different resource service providers can be classified according to resource types.
  • the resource types can be represented by different tags such as animation, music, dance, life, digital, and fashion.
  • the same resources may appear in different resource types. For example, for a certain music video resource, it can be divided into three resource types: animation, music, and fashion at the same time. Therefore, there may be duplicate resources in different resource servers.
  • these different resource servers can all be connected to the intelligent identification system.
  • the intelligent identification system can open a third-party interface, and at the same time, the resource server can also open its own interface to the intelligent identification system, and the intelligent identification system and the resource server can be communicated in accordance with existing or custom network communication protocols. Data interaction.
  • the intelligent identification system can be associated with multiple resource servers provided by multiple resource service providers.
  • the resource retrieval method in the intelligent identification system provided by the present application may include the following steps.
  • S1 Receive search information input by the user, and identify the resource type represented by the search information.
  • the intelligent identification system can receive the retrieval information input by the user through information entry components such as a microphone, a camera, a touch screen, and a keyboard.
  • the retrieval information input by the user can be any one of text, pictures, audio, and video, or a combination of any of them.
  • the intelligent identification system can identify the resource type represented by the retrieval information.
  • the intelligent identification system in addition to establishing associations with multiple resource servers, can also establish associations with multiple analysis servers provided by multiple analysis service providers. The analysis server can be used to analyze the search information input by the user, so as to extract the search keywords in the search information.
  • the types of analysis servers can also be diversified.
  • the multiple analysis servers may be divided according to the information type of the retrieved information.
  • the information type may be at least one of the above-mentioned text, picture, video, and audio.
  • some parsing servers can have multiple functions described above, and these parsing servers can be divided into multiple information types at the same time.
  • the intelligent identification system can identify the information type of the search information, and determine the target analysis server corresponding to the information type among the multiple analysis servers. Specifically, the intelligent identification system may, according to the identified information type, use the analysis server classified under the information type as the aforementioned target analysis server. Then, the intelligent recognition system may send the search information to the target analysis server to analyze the search keywords contained in the search information through the target analysis server. Finally, the intelligent recognition system can receive the search keywords fed back by the target analysis server.
  • the parsing server can perform the parsing process of the search keywords in different ways for the retrieval information of different information types. For example, if the user input is a text, the parsing server can analyze the semantics of the text through Natural Language Processing (NLP) technology, and extract the search keywords in the text based on the parsed semantics. For another example, if the user inputs a picture or video, the parsing server can analyze the image features in the picture or video frame according to the image semantic segmentation algorithm, and can use the extracted image features as search keywords. For another example, if the user input is voice, then the parsing server can use voice recognition technology to convert the voice into text, and then use NLP technology to analyze the semantics of the text, and extract the search keywords in the text based on the parsed semantics .
  • NLP Natural Language Processing
  • search keywords after the search keywords are identified, it is possible to determine which type of search resource the user is currently searching for based on the meaning of the search keywords.
  • different search keywords can have different type tags, and the resource types corresponding to the search keywords can be known according to the type tags of the search keywords. For example, if the search keyword "Like you slowly" has a music tag, then the user is currently searching for search resources of the music type. In this way, according to the retrieval information input by the user, the resource type characterized by the retrieval information can be determined.
  • S3 Determine several target resource servers corresponding to the resource type among the multiple resource servers, and send the retrieval information to the several target resource servers.
  • the intelligent recognition system can send the retrieval information or retrieval keywords extracted from the retrieval information to the several target resource servers.
  • S5 Receive retrieval resources fed back by the several target resource servers for the retrieval information, and show the received retrieval resources to the user.
  • each of the target resource servers after each of the target resource servers receives the search information or search keywords, they can search the resource database for search resources that match the search information or search keywords according to their own search strategy. , And feed retrieval resources to the intelligent identification system.
  • the intelligent identification system after the intelligent identification system receives the retrieval resource fed back by the target resource server, it can show the retrieval resource to the user. It should be noted that, for the same retrieval information, the intelligent identification system can receive retrieval resources fed back by several target resource servers. Therefore, there can also be multiple retrieval resources displayed to the user. In this way, users can select one or more of the retrieval resources to browse according to their own judgment and needs.
  • the target resource server can only provide a link to the search resource to the intelligent identification system.
  • the target resource server can directly provide the user with the search resource, which can avoid a large amount of data search Resources are frequently transmitted in the network, but are directly provided by the target resource server to the user's client.
  • the intelligent identification system can send a prompt message to the target resource server corresponding to the selected search resource, and the prompt message can include the network address of the user's client.
  • the target resource server can send the retrieval resource to the user's client according to the network address. That is, the intelligent recognition system can determine the target search resource selected by the user among the displayed search resources, and provide the target search resource to the user through the resource server corresponding to the target search resource.
  • the aforementioned user client, resource server, parsing server, and intelligent identification system can be accelerated through a Content Delivery Network (CDN).
  • CDN Content Delivery Network
  • the user's client and the intelligent identification system between the user's client and each of the resource servers, between the intelligent identification system and each of the resource servers, and between the intelligent identification system Between each analysis server, can be accelerated through the content distribution network.
  • the search information can be analyzed in real time while the user enters the search information.
  • the information fragment of the search information received at the current moment can be sent to the target analysis server, so as to analyze the partial information contained in the information fragment through the target analysis server.
  • Search keywords For example, the complete retrieval information that the user is going to enter is "I want to search for the title song of Jay Chou's latest album "Cowboy is busy"", then when the user enters "I want to search for Jay Chou", this piece of information can be sent directly to the target Analysis server, then the target analysis server can parse out that the partial search keyword is "Jay Chou".
  • the target analysis server can obtain various partial search keywords at different stages. For example, for the above search information, multiple partial search keywords such as “Jay Chou”, “Album”, “Title Song”, and “Cowboy is busy” can be obtained at different stages. Subsequently, the target parsing server may generate the retrieval keywords of the retrieval information based on each partial retrieval keyword. Specifically, the combination of these partial search keywords can be used as the final search keywords, or they can be sorted according to the weight values of these partial search keywords, and the partial search keywords with higher weight values can be filtered out, and the selected ones The partial search keyword is the most final search keyword.
  • a unified payment portal can be configured in the smart identification system. Through the unified payment portal, users can pay the smart identification system for resource retrieval fees without having to pay fees to various resource service providers. Then, in order to compensate the copyright fees of the resource service providers, the intelligent identification system can allocate the resource search fees or part of the resource search fees among multiple resource service providers according to the number of times the search resource is adopted by the user.
  • the intelligent recognition system can determine the target search resource selected by the user among the displayed search resources, and identify the target resource service provider corresponding to the target search resource, thereby The number of times that the retrieval resources provided by each resource service provider are adopted by users can be counted. For example, if a payment period is 7 days, then the intelligent identification system can count the total number of times that the resources provided by each resource service provider are adopted by users within 7 days. Then, the intelligent identification system can determine the respective cost sharing ratio of each resource service provider according to the number of times that the retrieval resource provided by each resource service provider is adopted by the user. The cost sharing ratio can be the ratio of the number of times the resource is adopted.
  • the intelligent identification system can receive the resource retrieval fee paid by the user in the intelligent identification system, and allocate the resource retrieval fee or part of the resource retrieval fee to each of the resource retrieval fees according to the fee sharing ratio.
  • Resource service provider there are currently 3 resource service providers, and the resources provided by these 3 resource service providers are adopted 500,000 times, 100,000 times, and 400,000 times, respectively. Then the cost sharing ratio of these 3 resource service providers can be It is 5:1:4, and the respective cost sharing ratios are 50%, 10% and 40%.
  • the intelligent identification system can receive the resource retrieval fee paid by the user in the intelligent identification system, and allocate the resource retrieval fee or part of the resource retrieval fee to each of the resource retrieval fees according to the fee sharing ratio.
  • the intelligent identification system can also count the number of analysis services provided by each analysis service provider within a specified time period, and determine the proportion of each analysis service provider according to the number of analysis services. In the end, the intelligent identification system can allocate part of the resource search costs among the analysis service providers according to the proportion of each analysis service provider.
  • the intelligent identification system may generate and store a unique identification code for marking this retrieval process after receiving the retrieval resources fed back by the several target resource servers for the retrieval information.
  • the unique identification code may be a hash value
  • the data used to calculate the hash value may be the analysis result obtained by the analysis server, the time required for the analysis, the retrieval resources fed back by the resource server, and the user selection Search resources, etc.
  • These data can be represented by binary or hexadecimal, and then a hash value is calculated through a hash algorithm, and the hash value can be used as the above-mentioned unique identification code.
  • a unique identification code can be generated during each retrieval process, based on the unique identification code and the analysis results involved in the retrieval process, the time required for analysis, the retrieval resources fed back by the resource server, and the retrieval resources selected by the user.
  • a block of the current retrieval process can be generated, wherein the unique identification code can be located in the head of the block, and the data involved in the retrieval process can be located in the body of the block.
  • different retrieval processes can correspond to different blocks, and each block can form a block chain and be stored on the Internet, so as to ensure that the data in the retrieval process is not easily tampered with, thereby providing a guarantee for subsequent profit sharing.
  • the unique identification code obtained by the above calculation the corresponding block can be queried and the data in the block body can be identified. According to these data, the resource search fees paid by the user in the intelligent identification system can be allocated among the target resource servers.
  • the intelligent identification system can evaluate the priority of the resource service provider according to the user's behavior. Specifically, the intelligent identification system may determine the resource display priority of each resource service provider according to the number of times that the retrieval resource provided by each resource service provider is adopted by the user. Among them, the more times the retrieved resource is adopted by users, the higher the priority of resource display.
  • the intelligent recognition system displays the search resources provided by each resource service provider to the user again, the displayed search resources can be sorted according to the resource display priority, and the search resources with higher resource display priority are given priority to the user Display, which can provide users with relatively high-quality retrieval resources.
  • the present application also provides an intelligent identification system that is associated with multiple resource servers provided by multiple resource service providers, and the intelligent identification system includes:
  • the retrieval information input unit is used to receive retrieval information input by the user and identify the resource type represented by the retrieval information
  • the resource search unit is configured to determine several target resource servers corresponding to the resource type among the multiple resource servers, and send the search information to the several target resource servers;
  • the resource display unit is configured to receive the retrieval resources fed back by the several target resource servers for the retrieval information, and display the received retrieval resources to the user.
  • the intelligent identification system is also associated with multiple analysis servers provided by multiple analysis service providers; accordingly, the intelligent identification system further includes:
  • the search information analysis unit is used to identify the information type of the search information, and determine the target analysis server corresponding to the information type among the multiple analysis servers; wherein, the information type includes text, picture, video, At least one of audio;
  • the keyword receiving unit is configured to send the search information to the target analysis server to analyze the search keywords contained in the search information through the target analysis server, and receive the feedback from the target analysis server Search keywords;
  • the search information entry unit is used to identify the resource type represented by the search keywords, and the resource search unit is used to send the search keywords to the plurality of target resource servers, and the The resource display unit is configured to receive the retrieval resources fed back by the plurality of target resource servers for the retrieval keyword.
  • the intelligent identification system further includes:
  • the adoption count counting unit is used to determine the target search resource selected by the user among the displayed search resources, and identify the target resource service provider corresponding to the target search resource, so as to count the searches provided by each resource service provider The number of times the resource has been adopted by users.
  • the intelligent identification system further includes:
  • the share ratio determining unit is configured to determine the respective cost share ratio of each resource service provider according to the number of times that the search resource provided by each resource service provider is adopted by the user;
  • the unified payment unit is used to receive the resource retrieval fee paid by the user in the intelligent identification system, and allocate the resource retrieval fee or part of the resource retrieval fee to each Resource service provider.
  • the intelligent identification system further includes:
  • the display priority determining unit is configured to determine the resource display priority of each resource service provider according to the number of times the retrieval resource provided by each resource service provider is adopted by the user;
  • the resource sorting unit is used to sort the displayed search resources according to the resource display priority when the search resources provided by each resource service provider are displayed to the user again.
  • the intelligent identification system further includes:
  • the identification code generating unit is used to generate and store a unique identification code used to mark this retrieval process, so that the resource retrieval fee paid by the user in the intelligent identification system is calculated according to the unique identification code. Assign in the target resource server.
  • between the user's client and the intelligent identification system, between the user's client and each of the resource servers, and between the intelligent identification system and each of the resource servers , Are accelerated through the content distribution network.
  • the user-oriented intelligent identification system may have an information entry module, a resource display module, and a payment module, where the information entry module can receive retrieval information input by the user, and the resource display The module can display each retrieval resource (or the link of the retrieval resource) to the user, and the payment module can receive the resource retrieval fee paid by the user.
  • the intelligent identification system there may be an analysis server interface, a resource server interface, a unique identification code calculation unit, a cost division calculation unit, a priority sorting unit, and a CDN acceleration interface.
  • the intelligent identification system can provide a unified search entry, resource display exit, and payment entry.
  • the analysis of search information in the background, the acquisition of search resources, the sharing of benefits, etc. are unaware to users. In this way, users can download
  • the software of the intelligent identification system can obtain the retrieval resources of the entire network, and there is no need to pay for different resources, which greatly simplifies the user's operation process and reduces the cost of the user.
  • the technical solution provided by this application can associate multiple resource servers provided by different resource service providers to the same intelligent identification system, so that the intelligent identification system can provide users with a unified search portal.
  • the intelligent recognition system can preferentially identify the resource type represented by the search information. For example, the smart recognition system can recognize whether the user wants to search for music, videos, or articles.
  • the intelligent identification system can deliver the retrieval information to the corresponding target resource server for processing. Specifically, assuming that the intelligent recognition system recognizes that the user currently wants to retrieve video resources, the retrieval information can be sent to the resource servers of service providers such as Youku Video, Tencent Video, and iQiyi Video for processing.
  • each target resource server After each target resource server processes and obtains the retrieval resources, it can feed the retrieval resources back to the intelligent identification system. In this way, the intelligent identification system can display these retrieval resources to the user for selection. It can be seen from the above that this application provides a unified intelligent identification system. Users do not need to download multiple different resource software to obtain the retrieval resources they are interested in, thereby reducing the user’s use cost and simplifying the user’s retrieval method. .
  • each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware.
  • the above technical solutions can be embodied in the form of software products, which can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., include a number of instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Accounting & Taxation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Finance (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Information Transfer Between Computers (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本申请公开了一种智能识别系统中的资源检索方法及智能识别系统，其中，所述智能识别系统与多个资源服务商提供的多个资源服务器相关联，所述方法包括：接收用户输入的检索信息，并识别所述检索信息表征的资源类型(S1)；在所述多个资源服务器中确定所述资源类型对应的若干个目标资源服务器，并将所述检索信息发送至所述若干个目标资源服务器处(S3)；接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源，并向所述用户展示接收到的所述检索资源(S5)。本申请提供的技术方案，能够简化用户的资源获取方式。

Description

一种智能识别系统中的资源检索方法及智能识别系统
交叉引用
本申请引用于2019年3月6日递交的名称为“一种智能识别系统中的资源检索方法及智能识别系统”的第201910169485.X号中国专利申请，其通过引用被全部并入本申请。
技术领域
本申请涉及互联网技术领域，特别涉及一种智能识别系统中的资源检索方法及智能识别系统。
背景技术
随着版权意识的不断增强，互联网中的各项资源也逐渐受到版权保护。对于不同的服务商而言，往往会针对部分资源进行版权保护。例如，在音乐版块，当前存在腾讯、阿里巴巴、网易等多个不同的音乐服务商，这些音乐服务商通常会推出自身的音乐软件，在音乐软件内可以搜索到用户感兴趣的音乐资源。而每个音乐服务商往往都会通过购买音乐版权，从而独占某些音乐资源。这部分独占的音乐资源，只能由对应的音乐服务商提供，在其它的音乐服务商的音乐软件内就无法搜索到对应内容。
针对上述的情形，用户为了能够获取到全部的资源，往往需要同时下载多个资源服务商的资源软件，通过在不同的资源软件之间切换，用户才能获取到自己感兴趣的资源。这样无疑会增加用户的资源获取成本，并且这样的资源获取方式不够便捷。
发明内容
本申请的目的在于提供一种智能识别系统中的资源检索方法及智能识别系统，能够简化用户的资源获取方式。
为实现上述目的，本申请一方面提供一种智能识别系统中的资源检索方法，所述智能识别系统与多个资源服务商提供的多个资源服务器相关联，所述方法包括：接收用户输入的检索信息，并识别所述检索信息表征的资源类型；在所述多个资源服务器中确定所述资源类型对应的若干个目标资源服务器，并将所述检索信息发送至所述若干个目标资源服务器处；接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源，并向所述用户展示接收到的所述检索资源。
为实现上述目的，本申请另一方面还提供一种智能识别系统，所述智能识别系统与多个资源服务商提供的多个资源服务器相关联，所述智能识别系统包括：检索信息录入单元，用于接收用户输入的检索信息，并识别所述检索信息表征的资源类型；资源检索单元，用于在所述多个资源服务器中确定所述资源类型对应的若干个目标资源服务器，并将所述检索信息发送至所述若干个目标资源服务器处；资源展示单元，用于接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源，并向所述用户展示接收到的所述检索资源。
由上可见，本申请提供的技术方案，可以将不同的资源服务商提供的多个资源服务器均关联至同一个智能识别系统中，这样，智能识别系统可以向用户提供统一的检索入口。用户在该检索入口中录入检索信息后，智能识别系统可以优先识别该检索信息表征的资源类型。例如，智能识别系统可以识别用户是想搜索音乐还是视频还是文章。在识别出资源类型后，智能识别系统可以将该检索信息下发至对应的目标资源服务器处进行处理。具体地，假设智能识别系统识别出用户当前想要检索视频资源，那么便可以将该检索信息发送至优酷视频、腾讯视频、爱奇艺视频等服务商的资源服务器处分别进行处理。当各个目标资源服务器处理得到检索资源后，便可以将检索资源反馈给智能识别系统， 这样，智能识别系统便可以将这些检索资源展示给用户，以供用户选择。由上可见，本申请提供了一个统一的智能识别系统，用户无需下载多个不同的资源软件，便可以获取到自己感兴趣的检索资源，从而减少了用户的使用成本，简化了用户的检索方式。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请实施例中资源检索方法的步骤示意图；
图2是本申请实施例中智能识别系统的网络架构示意图；
图3是本申请实施例中智能识别系统的功能模块示意图；
图4是本申请实施例中智能识别系统的内部组件示意图。
具体实施例
为使本申请的目的、技术方案和优点更加清楚，下面将结合附图对本申请实施例作进一步地详细描述。
本申请提供一种智能识别系统中的资源检索方法，所述智能识别系统可以与不同的资源服务商提供的多个资源服务器对接。这些不同的资源服务商提供的资源服务器可以按照资源类型进行分类。在实际应用中，所述资源类型可以利用动画、音乐、舞蹈、生活、数码、时尚等不同的标签来表示。需要说明的是，不同的资源类型中，可能会出现相同的资源。例如，对于某个音乐视频的资源而言，其可以同时被划分至动画、音乐、时尚这三种资源类型中。因此，不同的资源服务器内，可以存在重复的资源。并且，同一个资源类型下，也可以存在不同资源服务商提供的多个资源服务器。例如，在“音乐”这一资源类型下， 可以存在腾讯、阿里巴巴、网易这三个资源服务商提供的三个或者更多个资源服务器。
在本实施例中，对不同的资源服务器按照资源类型进行划分后，可以将这些不同的资源服务器都对接至所述智能识别系统中。具体地，所述智能识别系统可以开放第三方接口，同时，资源服务器也可以向智能识别系统开放自身的接口，智能识别系统与资源服务器之间可以按照现有的或者自定义的网络通信协议进行数据交互。这样，所述智能识别系统便可以与多个资源服务商提供的多个资源服务器相关联。
请参阅图1，本申请提供的智能识别系统中的资源检索方法可以包括以下步骤。
S1：接收用户输入的检索信息，并识别所述检索信息表征的资源类型。
在本实施例中，智能识别系统可以通过麦克风、摄像头、触控屏、键盘等信息录入组件接收用户输入的检索信息。相应地，用户输入的检索信息可以是文字、图片、音频、视频中的任意一种或者是任意多种的组合。在接收到用户输入的检索信息后，智能识别系统可以识别该检索信息表征的资源类型。具体地，请参阅图2，智能识别系统除了与多个资源服务器建立关联，还可以与多个解析服务商提供的多个解析服务器建立关联。所述解析服务器可以用于对用户输入的检索信息进行解析，从而提取出检索信息中的检索关键词。为了能够针对不同类型的检索信息进行解析，解析服务器的类型也可以多样化。所述多个解析服务器可以按照检索信息的信息类型进行划分。所述信息类型可以是上述的文字、图片、视频、音频中的至少一种。这样，按照信息类型划分之后，可以具备用于实现语义解析的解析服务器，还可以具备用于实现图像分析的解析服务器，还可以具备用于实现语音信息识别的解析服务器，以及具备用于实现视频画面分析的解析服务器。当然，在实际应用中，部分解析服务器可以具备上述的多个功能，这部分解析服务器可以被同时划分至多种信息类型中。
在本实施例中，智能识别系统在接收到用户输入的检索信息后，可以识 别所述检索信息的信息类型，并在所述多个解析服务器中确定出所述信息类型对应的目标解析服务器。具体地，智能识别系统可以根据识别出的信息类型，将划分至该信息类型下的解析服务器作为上述的目标解析服务器。然后，智能识别系统可以将所述检索信息发送至所述目标解析服务器，以通过所述目标解析服务器解析所述检索信息中包含的检索关键词。最后，智能识别系统便可以接收所述目标解析服务器反馈的所述检索关键词。
在实际应用中，针对不同信息类型的检索信息，解析服务器可以按照不同的方式进行检索关键词的解析过程。举例来说，若用户输入的是文字，那么解析服务器可以通过自然语言处理(Natural Language Processing，NLP)技术来解析文字的语义，并根据解析出的语义提取出文字中的检索关键词。又例如，若用户输入的是图片或者视频，解析服务器可以按照图像语义分割算法，解析出图片或者视频帧中的图像特征，并可以将提取出的图像特征作为检索关键词。再例如，若用户输入的是语音，那么解析服务器可以通过语音识别技术，将语音转换为文字，然后再通过NLP技术来解析文字的语义，并根据解析出的语义提取出文字中的检索关键词。
在本实施例中，在识别出检索关键词后，可以根据检索关键词的含义，确定出用户当前待检索的是哪一类型的检索资源。具体地，不同的检索关键词可以具备不同的类型标签，根据检索关键词的类型标签便可以获知该检索关键词对应的资源类型。例如，检索关键词“慢慢喜欢你”具备音乐标签，那么用户当前待检索的就是音乐类型的检索资源。这样，根据用户输入的检索信息，便可以确定出该检索信息表征的资源类型。
S3：在所述多个资源服务器中确定所述资源类型对应的若干个目标资源服务器，并将所述检索信息发送至所述若干个目标资源服务器处。
在本实施例中，在识别出检索信息表征的资源类型后，可以将其划分至该资源类型下的资源服务器，作为所述资源类型对应的若干个目标资源服务器。然后，智能识别系统可以将所述检索信息或者根据该检索信息提取出的检索关 键词发送至所述若干个目标资源服务器处。
S5：接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源，并向所述用户展示接收到的所述检索资源。
在本实施例中，各个所述目标资源服务器在接收到检索信息或者检索关键词后，便可以按照自身的检索策略，在资源库中检索与所述检索信息或者检索关键词相匹配的检索资源，并将检索资源反馈至所述智能识别系统。
在本实施例中，智能识别系统接收到目标资源服务器反馈的检索资源后，便可以向用户展示所述检索资源。需要说明的是，针对相同的检索信息而言，智能识别系统可以接收若干个目标资源服务器反馈的检索资源，因此，向用户展示的检索资源也可以是多个。这样，用户可以根据自身的判断和需求，选择其中的一个或者多个检索资源进行浏览。
在实际应用中，目标资源服务器可以仅仅向智能识别系统提供指向检索资源的链接，当该链接被用户触发时，可以直接由目标资源服务器向用户提供检索资源，这样可以避免数据量较大的检索资源在网络中频繁传输，而是直接由目标资源服务器提供给用户的客户端。具体地，当用户在智能识别系统中选择了一个检索资源的链接后，智能识别系统便可以向选中的检索资源对应的目标资源服务器发送提示信息，该提示信息中可以包含用户的客户端的网络地址。这样，目标资源服务器便可以根据该网络地址，向用户的客户端发送检索资源。也就是说，智能识别系统可以确定所述用户在展示的所述检索资源中选择的目标检索资源，并通过所述目标检索资源对应的资源服务器向所述用户提供所述目标检索资源。
在实际应用中，为了提高检索过程的响应速度，可以从多方面进行改进。一方面，针对上述的用户的客户端、资源服务器、解析服务器以及智能识别系统之间，可以通过内容分发网络(Content Delivery Network，CDN)进行加速。具体地，在用户的客户端与所述智能识别系统之间、所述用户的客户端与各个所述资源服务器之间、所述智能识别系统与各个所述资源服务器之间，以及智 能识别系统与各个解析服务器之间，均可以通过内容分发网络进行加速。
另一方面，可以在用户录入检索信息的过程中，实时地对检索信息进行解析。具体地，在接收用户输入的检索信息的过程中，可以将当前时刻接收到的检索信息的信息片段发送至所述目标解析服务器，以通过所述目标解析服务器解析所述信息片段中包含的局部检索关键词。例如，用户准备录入的完整检索信息是“我想搜索周杰伦最新专辑的主打歌《牛仔很忙》”，那么当用户录入至“我想搜索周杰伦”时，这一信息片段便可以直接发送至目标解析服务器，那么目标解析服务器可以解析出局部检索关键词是“周杰伦”。这样，随着检索信息的不断完善，当智能识别系统接收到用户输入的完整的检索信息后，所述目标解析服务器可以得到不同阶段的各个局部检索关键词。例如对于上述检索信息而言，在不同阶段可以得到“周杰伦”、“专辑”、“主打歌”、“牛仔很忙”等多个局部检索关键词。后续，目标解析服务器可以基于各个局部检索关键词生成所述检索信息的检索关键词。具体地，可以将这些局部检索关键词的组合作为最终的检索关键词，也可以按照这些局部检索关键词的权重值进行排序，筛选出权重值较高的局部检索关键词，并将筛选出的局部检索关键词最为最终的检索关键词。例如上述例子中，可以将权重值最高的“牛仔很忙”作为最终的检索关键词。这样，通过边录入边解析的方式，而不是在用户录入完整的检索信息后再进行解析的方式，可以提高检索信息的解析速度。
在一个实施例中，考虑到现有技术中，用户如果需要获取全网的资源，可能需要对多个不同的资源服务商支付费用，这样会增加用户投入的成本，也会增加用户的操作负担。鉴于此，在本实施例中，可以在智能识别系统中配置统一的支付入口，用户通过该统一的支付入口，可以向智能识别系统支付资源检索费用，而无需再向各个资源服务商支付费用。然后，为了补偿资源服务商的版权费用，智能识别系统可以根据检索资源被用户采纳的次数，将资源检索费用或者资源检索费用中的一部分在多个资源服务商之间分配。
具体地，在向所述用户展示检索资源后，智能识别系统可以确定所述用 户在展示的所述检索资源中选择的目标检索资源，并识别所述目标检索资源对应的目标资源服务商，从而可以统计各个所述资源服务商提供的检索资源被用户采纳的次数。例如，一个支付周期是7天，那么智能识别系统可以在7天内统计各个资源服务商提供的资源被用户采纳的总次数。然后，智能识别系统可以根据各个所述资源服务商提供的检索资源被用户采纳的次数，确定各个所述资源服务商各自的费用分成比例。该费用分成比例就可以是资源被采纳的次数的比例。例如，当前共有3个资源服务商，这3个资源服务商提供的资源被采纳的次数分别是50万次，10万次和40万次，那么这3个资源服务商的费用分成比例便可以是5:1:4，各自的费用分成比例分别是50％、10％和40％。这样，智能识别系统可以接收用户在所述智能识别系统中支付的资源检索费用，并按照所述费用分成比例，将所述资源检索费用或者所述资源检索费用中的部分费用分配给各个所述资源服务商。
当然，对于解析服务商提供的解析服务，智能识别系统也可以统计指定时间周期内各个解析服务商提供解析服务的次数，并按照解析服务的次数来确定各个解析服务商的分成比例。最终，智能识别系统可以按照各个解析服务商的分成比例，将部分资源检索费用在各个解析服务商之间进行分配。
在实际应用中，为了提高统计数据的安全性，可以采用区块链技术来生成每个检索过程的数据。具体地，智能识别系统在接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源后，可以生成并存储用于标记本次检索过程的唯一识别码。在实际应用中，所述唯一识别码可以是一个哈希值，用于计算该哈希值的数据可以是解析服务器得到的解析结果、解析所需的时长、资源服务器反馈的检索资源以及用户选择的检索资源等。这些数据可以通过二进制或者十六进制进行表示，然后通过哈希算法，计算得到一个哈希值，该哈希值便可以作为上述的唯一识别码。这样，每次的检索过程均可以生成一个唯一识别码，根据该唯一识别码和检索过程中涉及的解析结果、解析所需的时长、资源服务器反馈的检索资源以及用户选择的检索资源等数据，可以生成当前检 索过程的区块(block)，其中，唯一识别码可以位于区块的区块头(head)，检索过程中涉及的数据可以位于区块的区块体(body)。这样，不同的检索过程可以对应不同的区块，各个区块可以构成区块链存储于互联网中，从而可以保证检索过程的数据不易被篡改，从而为后续的利益分成提供保证。这样，依据上述计算得到的唯一识别码，便可以查询得到对应的区块，从而可以识别区块体中的数据。根据这些数据，可以将用户在所述智能识别系统中支付的资源检索费用，在各个所述目标资源服务器中进行分配。
在一个实施例中，由于不同的资源服务商提供的检索资源的质量或者准确程度各不相同，因此用户对于检索资源也会有不同的反馈，该反馈可以通过检索资源被用户采纳的次数来表示。在本实施例中，智能识别系统可以针对用户的行为，对资源服务商的优先级进行评定。具体地，智能识别系统可以根据各个所述资源服务商提供的检索资源被用户采纳的次数，确定各个所述资源服务商的资源展示优先级。其中，检索资源被用户采纳的次数越多，资源展示优先级则越高。那么当智能识别系统再次向用户展示各个所述资源服务商提供的检索资源时，可以将展示的检索资源按照所述资源展示优先级进行排序，将资源展示优先级较高的检索资源优先向用户展示，从而可以向用户提供比较优质的检索资源。
请参阅图3，本申请还提供一种智能识别系统，所述智能识别系统与多个资源服务商提供的多个资源服务器相关联，所述智能识别系统包括：
检索信息录入单元，用于接收用户输入的检索信息，并识别所述检索信息表征的资源类型；
资源检索单元，用于在所述多个资源服务器中确定所述资源类型对应的若干个目标资源服务器，并将所述检索信息发送至所述若干个目标资源服务器处；
资源展示单元，用于接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源，并向所述用户展示接收到的所述检索资源。
在一个实施例中，所述智能识别系统还与多个解析服务商提供的多个解析服务器相关联；相应地，所述智能识别系统还包括：
检索信息解析单元，用于识别所述检索信息的信息类型，并在所述多个解析服务器中确定出所述信息类型对应的目标解析服务器；其中，所述信息类型包括文字、图片、视频、音频中的至少一种；
关键词接收单元，用于将所述检索信息发送至所述目标解析服务器，以通过所述目标解析服务器解析所述检索信息中包含的检索关键词，并接收所述目标解析服务器反馈的所述检索关键词；
相应地，所述检索信息录入单元用于识别所述检索关键词表征的资源类型，以及所述资源检索单元用于将所述检索关键词发送至所述若干个目标资源服务器处，以及所述资源展示单元用于接收所述若干个目标资源服务器针对所述检索关键词反馈的检索资源。
在一个实施例中，所述智能识别系统还包括：
采纳次数统计单元，用于确定所述用户在展示的所述检索资源中选择的目标检索资源，并识别所述目标检索资源对应的目标资源服务商，以统计各个所述资源服务商提供的检索资源被用户采纳的次数。
在一个实施例中，所述智能识别系统还包括：
分成比例确定单元，用于根据各个所述资源服务商提供的检索资源被用户采纳的次数，确定各个所述资源服务商各自的费用分成比例；
统一付费单元，用于接收用户在所述智能识别系统中支付的资源检索费用，并按照所述费用分成比例，将所述资源检索费用或者所述资源检索费用中的部分费用分配给各个所述资源服务商。
在一个实施例中，所述智能识别系统还包括：
展示优先级确定单元，用于根据各个所述资源服务商提供的检索资源被用户采纳的次数，确定各个所述资源服务商的资源展示优先级；
资源排序单元，用于当再次向用户展示各个所述资源服务商提供的检索 资源时，将展示的检索资源按照所述资源展示优先级进行排序。
在一个实施例中，所述智能识别系统还包括：
识别码生成单元，用于生成并存储用于标记本次检索过程的唯一识别码，以使得依据所述唯一识别码，将用户在所述智能识别系统中支付的资源检索费用，在各个所述目标资源服务器中进行分配。
在一个实施例中，所述用户的客户端与所述智能识别系统之间、所述用户的客户端与各个所述资源服务器之间、以及所述智能识别系统与各个所述资源服务器之间，均通过内容分发网络进行加速。
具体地，请参阅图4，在一个应用示例中，所述智能识别系统中面向用户可以具备信息录入模块、资源展示模块以及付费模块，其中，信息录入模块可以接收用户输入的检索信息，资源展示模块则可以向用户展示各个检索资源(或者检索资源的链接)，付费模块则可以接收用户支付的资源检索费用。而在该智能识别系统的后台，可以具备解析服务器接口、资源服务器接口、唯一识别码计算单元、费用分成计算单元、优先级排序单元、CDN加速接口等。对于用户而言，智能识别系统可以提供统一的检索入口、资源展示出口以及付费入口，后台的检索信息的解析、检索资源的获取、利益分成等，对于用户是无感知的，这样，用户通过下载智能识别系统的软件，便可以获取到全网的检索资源，并无需针对不同的资源进行不同的付费，极大地简化了用户的操作流程，也减少了用户所投入的成本。
由上可见，本申请提供的技术方案，可以将不同的资源服务商提供的多个资源服务器均关联至同一个智能识别系统中，这样，智能识别系统可以向用户提供统一的检索入口。用户在该检索入口中录入检索信息后，智能识别系统可以优先识别该检索信息表征的资源类型。例如，智能识别系统可以识别用户是想搜索音乐还是视频还是文章。在识别出资源类型后，智能识别系统可以将该检索信息下发至对应的目标资源服务器处进行处理。具体地，假设智能识别系统识别出用户当前想要检索视频资源，那么便可以将该检索信息发送至优酷 视频、腾讯视频、爱奇艺视频等服务商的资源服务器处分别进行处理。当各个目标资源服务器处理得到检索资源后，便可以将检索资源反馈给智能识别系统，这样，智能识别系统便可以将这些检索资源展示给用户，以供用户选择。由上可见，本申请提供了一个统一的智能识别系统，用户无需下载多个不同的资源软件，便可以获取到自己感兴趣的检索资源，从而减少了用户的使用成本，简化了用户的检索方式。
通过以上的实施例的描述，本领域的技术人员可以清楚地了解到各实施例可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件来实现。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。
以上所述仅为本申请的较佳实施例，并不用以限制本申请，凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (16)

  1. 一种智能识别系统中的资源检索方法，其中，所述智能识别系统与多个资源服务商提供的多个资源服务器相关联，所述方法包括：
    接收用户输入的检索信息，并识别所述检索信息表征的资源类型；
    在所述多个资源服务器中确定所述资源类型对应的若干个目标资源服务器，并将所述检索信息发送至所述若干个目标资源服务器处；
    接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源，并向所述用户展示接收到的所述检索资源。
  2. 根据权利要求1所述的方法，其中，所述智能识别系统还与多个解析服务商提供的多个解析服务器相关联；相应地，在接收到用户输入的检索信息后，所述方法还包括：
    识别所述检索信息的信息类型，并在所述多个解析服务器中确定出所述信息类型对应的目标解析服务器；将所述检索信息发送至所述目标解析服务器，以通过所述目标解析服务器解析所述检索信息中包含的检索关键词，并接收所述目标解析服务器反馈的所述检索关键词；
    相应地，识别所述检索关键词表征的资源类型，以及将所述检索关键词发送至所述若干个目标资源服务器处，并接收所述若干个目标资源服务器针对所述检索关键词反馈的检索资源。
  3. 根据权利要求2所述的方法，其中，所述方法还包括：
    在接收用户输入的检索信息的过程中，将当前时刻接收到的检索信息的信息片段发送至所述目标解析服务器，以通过所述目标解析服务器解析所述信息片段中包含的局部检索关键词，直至接收到用户输入的完整的检索信息后，所述目标解析服务器基于各个局部检索关键词生成所述检索信息的检索关键词。
  4. 根据权利要求1所述的方法，其中，在向所述用户展示接收到的所述检索资源后，所述方法还包括：
    确定所述用户在展示的所述检索资源中选择的目标检索资源，并识别所述目标检索资源对应的目标资源服务商，以统计各个所述资源服务商提供的检索资源被用户采纳的次数。
  5. 根据权利要求4所述的方法，其中，所述方法还包括：
    根据各个所述资源服务商提供的检索资源被用户采纳的次数，确定各个所述资源服务商各自的费用分成比例；
    接收用户在所述智能识别系统中支付的资源检索费用，并按照所述费用分成比例，将所述资源检索费用或者所述资源检索费用中的部分费用分配给各个所述资源服务商。
  6. 根据权利要求4所述的方法，其中，所述方法还包括：
    根据各个所述资源服务商提供的检索资源被用户采纳的次数，确定各个所述资源服务商的资源展示优先级；
    当再次向用户展示各个所述资源服务商提供的检索资源时，将展示的检索资源按照所述资源展示优先级进行排序。
  7. 根据权利要求1或4所述的方法，其中，在接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源后，所述方法还包括：
    生成并存储用于标记本次检索过程的唯一识别码，以使得依据所述唯一识别码，将用户在所述智能识别系统中支付的资源检索费用，在各个所述目标资源服务器中进行分配。
  8. 根据权利要求1或4所述的方法，其中，在向所述用户展示接收到的所 述检索资源后，所述方法还包括：
    确定所述用户在展示的所述检索资源中选择的目标检索资源，并通过所述目标检索资源对应的资源服务器向所述用户提供所述目标检索资源。
  9. 根据权利要求1或4所述的方法，其中，所述用户的客户端与所述智能识别系统之间、所述用户的客户端与各个所述资源服务器之间、以及所述智能识别系统与各个所述资源服务器之间，均通过内容分发网络进行加速。
  10. 一种智能识别系统，其中，所述智能识别系统与多个资源服务商提供的多个资源服务器相关联，所述智能识别系统包括：
    检索信息录入单元，用于接收用户输入的检索信息，并识别所述检索信息表征的资源类型；
    资源检索单元，用于在所述多个资源服务器中确定所述资源类型对应的若干个目标资源服务器，并将所述检索信息发送至所述若干个目标资源服务器处；
    资源展示单元，用于接收所述若干个目标资源服务器针对所述检索信息反馈的检索资源，并向所述用户展示接收到的所述检索资源。
  11. 根据权利要求10所述的智能识别系统，其中，所述智能识别系统还与多个解析服务商提供的多个解析服务器相关联；相应地，所述智能识别系统还包括：
    检索信息解析单元，用于识别所述检索信息的信息类型，并在所述多个解析服务器中确定出所述信息类型对应的目标解析服务器；其中，所述信息类型包括文字、图片、视频、音频中的至少一种；
    关键词接收单元，用于将所述检索信息发送至所述目标解析服务器，以通过所述目标解析服务器解析所述检索信息中包含的检索关键词，并接收所述目标解析服务器反馈的所述检索关键词；
    相应地，所述检索信息录入单元用于识别所述检索关键词表征的资源类型，以及所述资源检索单元用于将所述检索关键词发送至所述若干个目标资源服务器处，以及所述资源展示单元用于接收所述若干个目标资源服务器针对所述检索关键词反馈的检索资源。
  12. 根据权利要求10所述的智能识别系统，其中，所述智能识别系统还包括：
    采纳次数统计单元，用于确定所述用户在展示的所述检索资源中选择的目标检索资源，并识别所述目标检索资源对应的目标资源服务商，以统计各个所述资源服务商提供的检索资源被用户采纳的次数。
  13. 根据权利要求12所述的智能识别系统，其中，所述智能识别系统还包括：
    分成比例确定单元，用于根据各个所述资源服务商提供的检索资源被用户采纳的次数，确定各个所述资源服务商各自的费用分成比例；
    统一付费单元，用于接收用户在所述智能识别系统中支付的资源检索费用，并按照所述费用分成比例，将所述资源检索费用或者所述资源检索费用中的部分费用分配给各个所述资源服务商。
  14. 根据权利要求12所述的智能识别系统，其中，所述智能识别系统还包括：
    展示优先级确定单元，用于根据各个所述资源服务商提供的检索资源被用户采纳的次数，确定各个所述资源服务商的资源展示优先级；
    资源排序单元，用于当再次向用户展示各个所述资源服务商提供的检索资源时，将展示的检索资源按照所述资源展示优先级进行排序。
  15. 根据权利要求10或12所述的智能识别系统，其中，所述智能识别系统还包括：
    识别码生成单元，用于生成并存储用于标记本次检索过程的唯一识别码，以使得依据所述唯一识别码，将用户在所述智能识别系统中支付的资源检索费用，在各个所述目标资源服务器中进行分配。
  16. 根据权利要求10或12所述的智能识别系统，其中，所述用户的客户端与所述智能识别系统之间、所述用户的客户端与各个所述资源服务器之间、以及所述智能识别系统与各个所述资源服务器之间，均通过内容分发网络进行加速。
PCT/CN2019/081535 2019-03-06 2019-04-04 一种智能识别系统中的资源检索方法及智能识别系统 WO2020177178A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP19820654.2A EP3734473A4 (en) 2019-03-06 2019-04-04 RESOURCE SEARCH PROCESS AND INTELLIGENT IDENTIFICATION SYSTEM IN AN INTELLIGENT IDENTIFICATION SYSTEM
US16/726,063 US11122308B2 (en) 2019-03-06 2019-12-23 Resource retrieval method in an intelligent identification system and intelligent identification system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910169485.X 2019-03-06
CN201910169485.XA CN109871489A (zh) 2019-03-06 2019-03-06 一种智能识别系统中的资源检索方法及智能识别系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/726,063 Continuation US11122308B2 (en) 2019-03-06 2019-12-23 Resource retrieval method in an intelligent identification system and intelligent identification system

Publications (1)

Publication Number Publication Date
WO2020177178A1 true WO2020177178A1 (zh) 2020-09-10

Family

ID=66919929

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/081535 WO2020177178A1 (zh) 2019-03-06 2019-04-04 一种智能识别系统中的资源检索方法及智能识别系统

Country Status (4)

Country Link
US (1) US11122308B2 (zh)
EP (1) EP3734473A4 (zh)
CN (1) CN109871489A (zh)
WO (1) WO2020177178A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110990598B (zh) * 2019-11-18 2020-11-27 北京声智科技有限公司 资源检索方法、装置、电子设备及计算机可读存储介质
CN112528127A (zh) * 2020-05-30 2021-03-19 山东工商学院 一种基于大数据的平面设计作品匹配度分析系统
CN114897674A (zh) * 2022-07-14 2022-08-12 天津美腾科技股份有限公司 物块识别处理方法、转换设备、智能干选机及存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101202882A (zh) * 2007-07-19 2008-06-18 深圳市同洲电子股份有限公司 一种媒体资源传输方法、系统及机顶盒
CN102447677A (zh) * 2010-09-30 2012-05-09 北大方正集团有限公司 资源访问控制方法、系统和设备
CN103678560A (zh) * 2013-12-06 2014-03-26 乐视网信息技术（北京）股份有限公司 多媒体资源纠错检索方法、多媒体资源服务器及系统

Family Cites Families (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7062561B1 (en) * 2000-05-23 2006-06-13 Richard Reisman Method and apparatus for utilizing the social usage learned from multi-user feedback to improve resource identity signifier mapping
US20080148192A1 (en) * 2006-12-15 2008-06-19 Iac Search & Media, Inc. Toolbox pagination
CN101072116B (zh) * 2007-04-28 2011-07-20 华为技术有限公司 业务选择方法、装置、系统及客户端应用服务器
CN101819574A (zh) * 2009-10-13 2010-09-01 腾讯科技（深圳）有限公司 一种搜索引擎系统和信息搜索方法
US9906838B2 (en) * 2010-07-12 2018-02-27 Time Warner Cable Enterprises Llc Apparatus and methods for content delivery and message exchange across multiple content delivery networks
US10417037B2 (en) * 2012-05-15 2019-09-17 Apple Inc. Systems and methods for integrating third party services with a digital assistant
CN102946554B (zh) * 2012-09-29 2016-06-15 合一网络技术(北京)有限公司 一种按照网络视频播放量进行计费分成的方法及其系统
CN104284211A (zh) * 2013-07-05 2015-01-14 上海帕科软件科技有限公司 视频集中分发收费系统
CN103716326B (zh) * 2013-12-31 2017-02-01 华为技术有限公司 一种资源访问方法及用户资源网关
CN103944994A (zh) * 2014-04-25 2014-07-23 中国联合网络通信集团有限公司 分布式资源获取方法及装置
CN105446971B (zh) * 2014-06-16 2019-03-01 华为技术有限公司 一种信息显示方法和装置
US10467269B2 (en) * 2015-02-13 2019-11-05 Samsung Electronics Co., Ltd. Accessing category-specific search servers
CN106612311A (zh) * 2015-10-23 2017-05-03 中兴通讯股份有限公司 数据推送方法及装置
CN105847320A (zh) * 2016-03-02 2016-08-10 乐视致新电子科技（天津）有限公司 一种资源获取方法和装置
CN108322495B (zh) * 2017-01-18 2021-07-06 阿里巴巴集团控股有限公司 资源访问请求的处理方法、装置和系统
CN108460060A (zh) * 2017-02-22 2018-08-28 合网络技术（北京）有限公司 视频推荐方法及装置
CN108647341A (zh) * 2018-05-14 2018-10-12 佛山市真觉网络科技有限公司 一种提高搜索引擎排名的方法
CN108718347B (zh) * 2018-05-18 2021-12-03 腾讯科技（深圳）有限公司 一种域名解析方法、系统、装置及存储介质
CN113407317A (zh) * 2018-09-11 2021-09-17 华为技术有限公司 一种资源调度方法及装置
CN109410022A (zh) * 2018-09-30 2019-03-01 深圳市元征科技股份有限公司 一种基于区块链技术的海外购物退税方法及相关设备

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101202882A (zh) * 2007-07-19 2008-06-18 深圳市同洲电子股份有限公司 一种媒体资源传输方法、系统及机顶盒
CN102447677A (zh) * 2010-09-30 2012-05-09 北大方正集团有限公司 资源访问控制方法、系统和设备
CN103678560A (zh) * 2013-12-06 2014-03-26 乐视网信息技术（北京）股份有限公司 多媒体资源纠错检索方法、多媒体资源服务器及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3734473A4 *

Also Published As

Publication number Publication date
EP3734473A1 (en) 2020-11-04
EP3734473A4 (en) 2020-11-04
CN109871489A (zh) 2019-06-11
US11122308B2 (en) 2021-09-14
US20200329267A1 (en) 2020-10-15

Similar Documents

Publication Publication Date Title
US10832287B2 (en) Promotional content targeting based on recognized audio
CN107844586B (zh) 新闻推荐方法和装置
CN107832434B (zh) 基于语音交互生成多媒体播放列表的方法和装置
WO2020177178A1 (zh) 一种智能识别系统中的资源检索方法及智能识别系统
US11722575B2 (en) Dynamic application content analysis
CN111988663A (zh) 视频播放节点的定位方法、装置、设备以及存储介质
EP1810182A2 (en) Method and system for providing information services relevant to visual imagery
WO2009140828A1 (zh) 生成流媒体增值描述文件及插播多媒体信息的方法、系统
WO2022042157A1 (zh) 视频数据的制作方法、装置、计算机设备和存储介质
CN109063200B (zh) 资源搜索方法及其装置、电子设备、计算机可读介质
CN111597446A (zh) 基于人工智能的内容推送方法、装置、服务器和存储介质
US10909100B2 (en) Object identifier index
CN107729486B (zh) 一种视频搜索方法及装置
CN110035298B (zh) 一种媒体快速播放方法
CN106570003B (zh) 数据推送方法及装置
KR101840422B1 (ko) 사물 인식을 이용한 대상물 검색 시스템
WO2015012714A1 (ru) Способ проверки веб-страниц на содержание в них целевого аудио и/или видео (av) контента реального времени
JP2002358458A (ja) 地域情報配信システム及び地域情報配信プログラム
CN110753248B (zh) 一种用于iptv用户端数据收集系统及方法
CN114048348B (zh) 视频质量评分方法和装置、以及存储介质和电子设备
WO2023155502A1 (zh) 一种搜索内容分享方法、装置、计算机设备和存储介质
CN115955577A (zh) 评论处理方法、装置、计算设备及计算机存储介质
CN115299069A (zh) 数字视频分析

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2019820654

Country of ref document: EP

Effective date: 20191223

NENP Non-entry into the national phase

Ref country code: DE