為克服相關技術中存在的問題，本說明書提供了離線即時需求處理方法、資訊推薦方法、裝置及設備。
根據本說明書實施例的第一態樣，提供一種資訊推薦方法，所述方法包括：
對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，所述需求請求至少攜帶需求內容和需求方位置資訊；
至少根據所述意圖和所述需求方位置資訊對需求回應方進行初步篩選，獲得至少一個潛在需求回應方，並基於所述需求內容向潛在需求回應方推送需求提出方的需求；
基於潛在需求回應方反饋的回應資訊，從篩選獲得的潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方，並向需求提出方推薦所述目標需求回應方的相關資訊。
可選地，需求回應方和需求內容均按預設類別進行分類，所述意圖為所述需求內容對應的預設類別。
可選地，所述方法還包括：
從已知預設類別的需求回應方的資訊中，提取能表徵需求回應方所屬預設類別的特徵資料；
根據已知預設類別的需求回應方所屬預設類別和特徵資料，建構得到類別預測模型；
透過類別預測模型預測未知預設類別的需求回應方所屬預設類別。
可選地，所述特徵資料至少包括預設時間段內收款頻率、預設時間段內收款金額分佈、預設時間段內收款時間分佈和所處地理位置中的一種或多種特徵。
可選地，需求提出方的需求至少包括：基於所述需求內容獲得的音訊資料，所述音訊資料在潛在需求回應方被即時播放。
可選地，所述目標需求回應方的相關資訊包括以下一種或多種資訊：
目標需求回應方的標識資訊、目標需求回應方反饋的答覆內容、目標需求回應方與需求提出方的距離資訊、需求提出方所處位置到達目標需求回應方所處位置的導航指引資料。
可選地，向需求提出方推薦所述目標需求回應方包括：
將所述目標需求回應方的相關資訊，標記在地圖上與目標需求回應方的位置資訊相對應位置處，並向需求提出方發送已標記地圖。
根據本說明書實施例的第二態樣，提供一種離線即時需求處理方法，所述方法包括：
用戶端向服務端發送至少攜帶需求內容和需求方位置資訊的離線需求請求；
服務端對所述需求內容進行語義分析獲得用戶端的意圖，至少根據所述意圖和所述需求方位置資訊對商戶進行初步篩選，獲得至少一個潛在商戶，並基於所述需求內容向潛在商戶的終端推送用於表示用戶端需求的音訊資料；
潛在商戶的終端播放所述音訊資料；
服務端基於潛在商戶的終端播放所述音訊資料後反饋的回應資訊，從篩選獲得的潛在商戶中篩選出滿足用戶端需求的目標商戶，並向用戶端推薦所述目標商戶的相關資訊。
根據本說明書實施例的第三態樣，提供一種資訊推薦裝置，所述裝置包括：
意圖識別模組，用以：對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，所述需求請求至少攜帶需求內容和需求方位置資訊；
初步篩選模組，用以：至少根據所述意圖和所述需求方位置資訊對需求回應方進行初步篩選，獲得至少一個潛在需求回應方；
資訊傳輸模組，用以：基於所述需求內容向潛在需求回應方推送需求提出方的需求；
目標篩選模組，用以：基於潛在需求回應方反饋的回應資訊，從篩選獲得的潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方；
所述資訊傳輸模組，還用以向需求提出方推薦所述目標需求回應方的相關資訊。
可選地，需求回應方和需求內容均按預設類別進行分類，所述意圖為所述需求內容對應的預設類別。
可選地，所述裝置還包括：
模型訓練模組，用以：從已知預設類別的需求回應方的資訊中，提取能表徵需求回應方所屬預設類別的特徵資料；根據已知預設類別的需求回應方所屬預設類別和特徵資料，建構得到類別預測模型；
類別預測模組，用以：透過類別預測模型預測未知預設類別的需求回應方所屬預設類別。
可選地，所述特徵資料至少包括預設時間段內收款頻率、預設時間段內收款金額分佈、預設時間段內收款時間分佈和所處地理位置中的一種或多種特徵。
可選地，需求提出方的需求至少包括：基於所述需求內容獲得的音訊資料，所述音訊資料在潛在需求回應方被即時播放。
可選地，所述目標需求回應方的相關資訊包括以下一種或多種資訊：
目標需求回應方的標識資訊、目標需求回應方反饋的答覆內容、目標需求回應方與需求提出方的距離資訊、需求提出方所處位置到達目標需求回應方所處位置的導航指引資料。
可選地，所述裝置還包括：
資訊標記模組，用以：將所述目標需求回應方的相關資訊，標記在地圖上與目標需求回應方的位置資訊相對應位置處；
所述資訊傳輸模組用以向需求提出方發送已標記地圖。
根據本說明書實施例的第四態樣，提供一種電腦設備，包括記憶體、處理器及儲存在記憶體上並可在處理器上運行的電腦程式，其中，所述處理器執行所述程式時實現如下方法：
對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，所述需求請求至少攜帶需求內容和需求方位置資訊；
至少根據所述意圖和所述需求方位置資訊對需求回應方進行初步篩選，獲得至少一個潛在需求回應方，並基於所述需求內容向潛在需求回應方推送需求提出方的需求；
基於潛在需求回應方反饋的回應資訊，從篩選獲得的潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方，並向需求提出方推薦所述目標需求回應方的相關資訊。
本說明書的實施例，透過對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，由於需求請求還攜帶需求方位置資訊，因此，至少根據意圖和需求方位置資訊對需求回應方進行初步篩選，獲得至少一個潛在需求回應方，並基於需求內容向潛在需求回應方推送需求提出方的需求。基於潛在需求回應方反饋的回應資訊，從篩選獲得的潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方，並向需求提出方推薦所述目標需求回應方的相關資訊，從而實現提供一種需求提出方和需求回應方的互動通道，以解決雙方不能有效撮合、溝通的問題，便於用戶快速查到需要的需求回應方，進而提高用戶獲得服務或商品的效率。
應當理解的是，以上的一般描述和後文的細節描述僅是示例性和解釋性的，並不能限制本說明書。In order to overcome the problems in the related technologies, this specification provides an offline real-time demand processing method, information recommendation method, device and equipment.
According to the first aspect of the embodiments of the present specification, an information recommendation method is provided. The method includes:
Perform semantic analysis on the demand content in the demand request sent by the demand proposer to obtain the intention of the demand proposer. The demand request carries at least the demand content and the demand side location information;
At least preliminary screening the demand responders based on the intention and the demand party location information, obtaining at least one potential demand responder, and pushing the demand proposer's demand to the potential demand responder based on the demand content;
Based on the response information fed back by the potential demand responders, a target demand responder that meets the needs of the demand proposer is selected from the potential demand responders obtained by the screening, and the relevant information of the target demand responder is recommended to the demand proposer.
Optionally, the demand responder and the demand content are classified according to preset categories, and the intention is the preset category corresponding to the demand content.
Optionally, the method further includes:
Extract the characteristic data that can represent the default category to which the demand responder belongs from the information of the demand responder of the known preset category;
According to the preset categories and characteristic data of the demand responders of known preset categories, construct a category prediction model;
Use the category prediction model to predict the default category to which the demand responders of unknown preset categories belong.
Optionally, the characteristic data includes at least one or more characteristics of a collection frequency within a preset time period, a distribution of payment amount within a preset time period, a distribution time of payment within a preset time period, and a geographical location.
Optionally, the demand of the demand proposer includes at least: audio data obtained based on the content of the demand, and the audio data is played in real time on the potential demand responder.
Optionally, the relevant information of the target demand responder includes one or more of the following information:
The identification information of the target demand responder, the response content of the target demand responder feedback, the distance information between the target demand responder and the demand proposer, the navigation guidance data of the location of the demand proposer to reach the location of the target demand responder.
Optionally, recommending the target demand responder to the demand proposer includes:
Mark the relevant information of the target demand responder on the map at a position corresponding to the position information of the target demand responder, and send the marked map to the demand proposer.
According to a second aspect of the embodiments of the present specification, an offline instantaneous demand processing method is provided. The method includes:
The user end sends an offline demand request that carries at least demand content and demand-side location information to the server;
The server performs semantic analysis on the demand content to obtain the user's intention, at least preliminary screening the merchant according to the intention and the demand side location information, obtaining at least one potential merchant, and based on the demand content to the potential merchant's terminal Push the audio data used to represent the client's needs;
The terminal of the potential merchant plays the audio data;
Based on the response information fed back by the terminal of the potential merchant after playing the audio data, the server selects target merchants that meet the needs of the user from the potential merchants obtained by screening, and recommends the relevant information of the target merchant to the client.
According to a third aspect of the embodiments of the present specification, an information recommendation device is provided. The device includes:
Intent recognition module, used to: semantically analyze the demand content in the demand request sent by the demand proposer to obtain the intention of the demand proposer, the demand request carries at least the demand content and the demand side location information;
The preliminary screening module is used to: initially screen the demand responders based on the intention and the demand party location information to obtain at least one potential demand responder;
The information transmission module is used to: push the demand requester's demand to potential demand responders based on the demand content;
The target screening module is used to: based on the response information fed back by the potential demand responders, select the target demand responders that meet the needs of the demand proposer from the potential demand responders obtained by the screening;
The information transmission module is also used to recommend the relevant information of the target demand responder to the demand proposer.
Optionally, the demand responder and the demand content are classified according to preset categories, and the intention is the preset category corresponding to the demand content.
Optionally, the device further includes:
The model training module is used to: extract the characteristic data that can represent the preset category of the demand responder from the information of the demand responder of the known preset category; the preset category of the demand responder according to the known preset category And feature data to construct a category prediction model;
The category prediction module is used to predict the default category to which the demand responder of the unknown preset category belongs through the category prediction model.
Optionally, the characteristic data includes at least one or more characteristics of a collection frequency within a preset time period, a distribution of payment amount within a preset time period, a distribution time of payment within a preset time period, and a geographical location.
Optionally, the demand of the demand proposer includes at least: audio data obtained based on the content of the demand, and the audio data is played in real time on the potential demand responder.
Optionally, the relevant information of the target demand responder includes one or more of the following information:
The identification information of the target demand responder, the response content of the target demand responder feedback, the distance information between the target demand responder and the demand proposer, the navigation guidance data of the location of the demand proposer to reach the location of the target demand responder.
Optionally, the device further includes:
The information marking module is used to mark the relevant information of the target demand responder on the map at a position corresponding to the position information of the target demand responder;
The information transmission module is used to send the marked map to the requester.
According to a fourth aspect of the embodiments of the present specification, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program Implement the following method:
Perform semantic analysis on the demand content in the demand request sent by the demand proposer to obtain the intention of the demand proposer. The demand request carries at least the demand content and the demand side location information;
At least preliminary screening the demand responders based on the intention and the demand party location information, obtaining at least one potential demand responder, and pushing the demand proposer's demand to the potential demand responder based on the demand content;
Based on the response information fed back by the potential demand responders, a target demand responder that meets the needs of the demand proposer is selected from the potential demand responders obtained by the screening, and the relevant information of the target demand responder is recommended to the demand proposer.
In the embodiments of the present specification, the semantics of the demand content in the demand request sent by the demand requester are analyzed to obtain the intention of the demand requester. Since the demand request also carries the location information of the demander, at least according to the intention and the location information of the demander Conduct preliminary screening of demand responders, obtain at least one potential demand responder, and push the demand proposer's demand to potential demand responders based on the content of the demand. Based on the response information fed back by the potential demand responders, the target demand responders satisfying the needs of the demand proposer are selected from the potential demand responders obtained by the screening, and the relevant information of the target demand responders is recommended to the demand proposer to achieve Provide an interactive channel between the demand proposer and the demand responder to solve the problem that the two parties cannot effectively match and communicate, so that the user can quickly find the demand responder, thereby improving the user's efficiency in obtaining services or products.
It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit this specification.
這裡將詳細地對示例性實施例進行說明，其示例表示在圖式中。下面的描述涉及圖式時，除非另有表示，不同圖式中的相同數字表示相同或相似的要素。以下示例性實施例中所描述的實施方式並不代表與本說明書相一致的所有實施方式。相反，它們僅是與如所附申請專利範圍中所詳述的、本說明書的一些態樣相一致的裝置和方法的例子。
在本說明書使用的術語是僅僅出於描述特定實施例的目的，而非旨在限制本說明書。在本說明書和所附申請專利範圍中所使用的單數形式的“一種”、“所述”和“該”也旨在包括多數形式，除非上下文清楚地表示其他含義。還應當理解，本文中使用的術語“和/或”是指並包含一個或多個相關聯的列出專案的任何或所有可能組合。
應當理解，儘管在本說明書可能採用術語第一、第二、第三等來描述各種資訊，但這些資訊不應限於這些術語。這些術語僅用來將同一類型的資訊彼此區分開。例如，在不脫離本說明書範圍的情況下，第一資訊也可以被稱為第二資訊，類似地，第二資訊也可以被稱為第一資訊。取決於語境，如在此所使用的詞語“如果”可以被解釋成為“在……時”或“當……時”或“回應於確定”。
在實際應用中，當用戶提出某種需求時，往往很難確定哪些需求回應方能回應該需求，往往需要實地考察以確認需求回應方是否能提供該需求對應的服務或商品，可見，用戶獲得服務或商品的效率低。
有鑒於此，本說明書實施例提供一種需求提出方和需求回應方的互動通道，以解決雙方不能有效撮合、溝通的問題，進而提高用戶獲得服務或商品的效率。
下面將結合本說明書實施例中的圖式，對本說明書實施例中的技術方案進行示例說明。
如圖1所示，是本說明書根據一示例性實施例示出的一種資訊推薦系統架構示意圖。在該示意圖中，可以包括需求提出端、服務端、需求回應端。需求提出端可以是提出需求的一端，例如可以是用戶端。需求提出端可以是能提供需求提出服務的應用程式，例如支付寶等應用程式。需求提出端也可以是具有需求提出功能的電子設備。電子設備可以是行動電話或其它手持可攜式設備，也可以是諸如腕錶設備、吊墜設備等稍微更小的可攜式設備，或者小型化設備、平板電腦、筆記型電腦、桌上型電腦、整合於電腦顯示器中的電腦或其它的電子裝備。服務端可以是多台伺服器設備的統稱，也可以是安裝在伺服器設備上的軟體的統稱。需求回應端可以是能回應相應需求的一端，例如，可以是商戶端。需求回應端可以是能提供需求回應服務的應用程式，也可以是具有需求回應功能的電子設備。
接著從服務端的角度對本說明書實施例進行示例說明。如圖2所示，是本說明書根據一示例性實施例示出的一種資訊推薦方法的流程圖，所述方法包括：
在步驟202中，對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，所述需求請求至少攜帶需求內容和需求方位置資訊；
在步驟204中，至少根據所述意圖和所述需求方位置資訊對需求回應方進行初步篩選，獲得至少一個潛在需求回應方，並基於所述需求內容向潛在需求回應方推送需求提出方的需求；
在步驟206中，基於潛在需求回應方反饋的回應資訊，從篩選獲得的潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方，並向需求提出方推薦所述目標需求回應方的相關資訊。
其中，需求提出方可以是用戶，也可以是用戶端。需求回應方可以是商戶，也可以是商戶端。需求內容用來描述需求提出方的需求，可以是由需求提出方發出的語音內容，也可以是文本內容。需求方位置資訊可以是需求方的地理位置，也可以是用來確定需求方地理位置的資訊，例如WiFi資訊等。
對需求提出方發送的需求內容進行語義分析時，若需求內容為語音內容，則可以先將需求內容進行語音辨識以解析成文本內容，並對文本內容進行語義分析。對需求內容進行語義分析，目的是為了識別需求提出方的意圖。在一個實施例中，可以將需求內容對應的預設類別作為需求提出方的意圖。例如，可以預先建構預設類別，預設類別也可以稱為類目體系或預設類型。需求回應方和需求內容均按預設類別進行分類。預設類別可以基於商戶類型進行劃分獲得，也可以基於用戶意圖進行劃分獲得。例如，預設類別可以包括：百貨類、食品類、水果類。
在該實施例中，可以將預設類別作為需求提出方的意圖，即對需求內容進行語義分析的目的是為了獲得需求內容所對應的預設類別，從而可以實現快速識別需求提出方的意圖。透過預設類別將需求內容和需求回應方進行關聯，可以根據需求內容所對應的預設類別初步篩選需求回應方。
語義分析的方法有很多種，以下列舉兩種分析方法進行示例說明。
在一個例子中，可以採用關鍵字匹配的方式獲得需求提出方的意圖。例如，對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，可以包括：
從所述需求內容中提取與預設對應關係中關鍵字匹配的關鍵字，所述預設對應關係是預先建構的關鍵字與預設類別的對應關係；
根據所述預設對應關係，將所匹配關鍵字對應的預設類別作為需求提出方的意圖。
在該實施例中，可以預先建構關鍵字與預設類別的對應關係。對應關係可以基於大資料分析而獲得。例如，可以根據用戶在電子商務平臺上搜尋商品的歷史記錄中搜尋詞與最終商品所屬預設類別的對應關係等獲得。本實施例利用關鍵字匹配的方式，將需求內容中匹配到的關鍵字對應的預設類別作為需求提出方的意圖，可以實現快速獲得需求提出方的意圖。
在另一個實施例中，可以將需求內容輸入預建構的意圖識別模型，獲得需求內容所對應的預設類別。其中，意圖識別模型可以是預先建構的用來識別用戶意圖的模型。例如，可以利用已知預設類別的需求內容建構訓練樣本，並利用訓練樣本對深度學習演算法進行訓練，獲得意圖識別模型。本實施例利用意圖識別模型預測需求內容對應的預設類別，從而實現對更多需求內容的預設類別進行識別。
需求回應方按預設類別進行分類。實際應用中，可能僅有部分需求回應方可以獲得其預設類別，而部分需求回應方未知其預設類別。例如，部分需求回應方所屬預設類別可以基於需求回應方上傳的預設類別而獲得，也可以基於電子商務平臺中需求回應方的屬性資訊而獲得。而針對不明確所屬預設類別的其他需求回應方，可以利用已知所屬預設類別的需求回應方預測未知所屬預設類別的需求回應方所屬類別。例如，在一個實施例中，所述方法還包括：
從已知預設類別的需求回應方的資訊中，提取能表徵需求回應方所屬預設類別的特徵資料；
根據已知預設類別的需求回應方所屬預設類別和特徵資料，建構得到類別預測模型；
透過類別預測模型預測未知預設類別的需求回應方所屬預設類別。
其中，需求回應方的資訊可以包括歷史收款記錄、靜態屬性資訊等與需求回應方相關的資訊。特徵資料可以是能表徵需求回應方所屬預設類別的特徵資料。在一個例子中，所述特徵資料至少包括預設時間段內收款頻率、預設時間段內收款金額分佈、預設時間段內收款時間分佈和所處地理位置中的一種或多種特徵。可見，可以透過收款頻率、收款金額、收款時間、所處位置等資訊反映需求回應方所屬預設類別，從而準確預測其他需求回應方所屬預設類別。應當理解的是，特徵資料還可以包括其他特徵資料，只要能表徵需求回應方所屬預設類別即可，在此不一一贅述。
根據已知預設類別的需求回應方所屬預設類別以及特徵資料，可以建構得到類別預測模型。例如，根據已知預設類別的需求回應方所屬預設類別以及特徵資料，對監督演算法進行訓練，獲得用來預測預設類別的類別預測模型。監督演算法可以是線性演算法、Logistic回歸，隨機森林等。
在該實施例中，透過上述方法可以獲得更多需求回應方所屬預設類別，從而拓寬可推薦的需求回應方。
進一步地，為了提高安全性，本說明書實施例所指需求回應方可以是滿足可靠性條件的回應方。例如，根據歷史收款記錄等歷史資訊對回應方進行篩選，以確保獲得的需求回應方為可靠的回應方。特別是，需求回應方為可靠的商戶。
在應用階段，獲得需求提出方的意圖後，可以根據預設篩選條件對需求回應方進行初步篩選，獲得至少一個潛在需求回應方。其中，預設篩選條件的篩選因數至少包括意圖。為了推薦與需求提出方位置具有關聯的資訊，預設篩選條件的篩選因數還包括需求方位置資訊。利用意圖對需求回應方進行初步篩選，可以是篩選出與意圖所屬預設類別相同的需求回應方。利用需求方位置資訊對需求回應方進行初步篩選，可以是篩選出與需求提出方在距離上存在關聯的需求回應方。例如，篩選出與需求提出方距離在預設範圍內的需求回應方，或者，篩選出與需求提出方歸屬於同一區域的需求回應方等。可以理解的是，還可以設定其他篩選條件，以篩選出適合向需求提出方推薦的潛在需求回應方。例如，預設篩選條件還可以包括：潛在需求回應方為所持終端具有音訊播放功能的需求回應方，以便潛在需求回應方的終端即時播放需求提出方的需求。如，潛在需求回應方的終端可以是支付寶盒子或其他同類型具有音箱功能的設備。
本實施例中，為了區分不同需求回應方，將初步篩選獲得的需求回應方稱為潛在需求回應方。透過初步篩選，可以減少服務端通知需求回應方的數量，同時降低對不相關的需求回應方的打擾。
在獲得潛在需求回應方後，可以基於需求內容向潛在需求回應方推送需求提出方的需求。在一個例子，可以直接將需求內容推送至需求回應方，以實現建立需求提出方和需求回應方的互動通道，便於雙方進行溝通。在另一個例子中，為了增加對需求提出方的需求進行提醒的力度，向需求回應方推送的需求提出方的需求至少包括基於需求內容獲得的音訊資料，以便潛在需求回應方即時播放音訊資料，便於即時需求被回應。具體地，若所述需求內容為語音資料，則直接將語音資料推送至潛在需求回應方；若需求內容為文本資料，則將需求內容透過語音合成而產生音訊資料，將音訊資料推送至各潛在需求回應方。
可見，透過將需求回應方的需求以音訊資料的方式在潛在需求回應方即時被播放，可以及時提醒潛在回應方，便於該潛在需求回應方及時做出反饋，實現需求提出方發出的即時需求能被及時反饋。
潛在需求回應方的設備對需求提出方的需求進行提醒後，如果商戶滿足該需求，可以透過潛在需求回應方回應該需求。回應的類型很多，例如，可以透過物理按鍵或者按鍵組合等方式，反饋是否能滿足需求提出方需求；又或者，透過文本或語音等方式的答覆內容反饋是否能滿足需求提出方需求。
在服務端接收到潛在需求回應方反饋的回應資訊後，可以從潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方，並向需求提出方推薦所述目標需求回應方的相關資訊。
關於基於回應資訊進行目標需求回應方的篩選，在一個實施例中，能滿足需求提出方需求的需求回應方才會發起回應資訊，因此，可以根據是否接收到回應資訊以確定該潛在需求回應方是否滿足需求提出方需求。具體地，可以將發起回應資訊的潛在需求回應方作為目標需求回應方。若沒有接收到潛在需求回應方反饋的回應資訊，可以預設為該潛在需求回應方不滿足需求提出方需求。在另一個實施例中，回應資訊可以包括潛在需求回應方的商家反饋的答覆內容，例如，語音內容或文本內容等。一方面，可以根據答覆內容分析潛在需求回應方是否能滿足需求提出方需求，從而將滿足需求的潛在需求回應方作為目標需求回應方。另一方面，可以預設認為發起回應資訊的潛在需求回應方能滿足需求提出方的需求，因此，直接將發起回應資訊的潛在需求回應方作為目標需求回應方，並將答覆內容作為目標需求回應方的相關資訊之一，推送至需求提出方。
向需求提出方推薦目標需求回應方的相關資訊，可以是為了讓需求提出方的用戶能瞭解目標需求回應方。有鑒於此，在一個例子中，所述目標需求回應方的相關資訊包括以下一種或多種資訊：
目標需求回應方的標識資訊、目標需求回應方反饋的答覆內容、目標需求回應方與需求提出方的距離資訊、需求提出方所處位置到達目標需求回應方所處位置的導航指引資料。
其中，目標需求回應方的標識資訊可以是商戶名稱等唯一標識商戶的標識。目標需求回應方的位置資訊可以根據回應中的GPS資訊、WiFi資訊等而確定，或者根據標識資訊查找目標需求回應方登記過的位置資訊而獲得等。
可見，透過展示目標需求回應方的相關資訊，可以避免用戶再次查詢其他資訊，減少用戶操作步驟。特別是，當相關資訊包括目標需求回應方反饋的答覆內容，可以實現需求提出方和需求回應方的通信。
進一步地，向需求提出方推薦所述目標需求回應方包括：將所述目標需求回應方的相關資訊，標記在地圖上與目標需求回應方的位置資訊相對應位置處，並向需求提出方發送已標記地圖。
可見，透過向需求提出方發送已標記地圖，可以實現在需求提出方利用地圖的方式展示目標需求回應方的相關資訊。如果商戶反饋的是語音資料，還可以是在地圖相應位置展示觸發語音播放的控制項，實現需求提出方和需求回應方的互動。
以上實施方式中的各種技術特徵可以任意進行組合，只要特徵之間的組合不存在衝突或矛盾，但是限於篇幅，並未進行一一描述，因此上述實施方式中的各種技術特徵的任意進行組合也屬於本說明書揭示的範圍。
以下結合一個具體的應用場景對本說明書實施例進行示例說明。
如圖3A所示，是本說明書根據一示例性實施例示出的一種離線即時需求處理方法的流程圖，所述方法包括：
用戶端向服務端發送至少攜帶需求內容和需求方位置資訊的離線需求請求(步驟302)。其中，離線需求請求，可以指該需求為離線需求、且需要即時被處理的請求。
服務端對所述需求內容進行語義分析獲得用戶端的意圖，至少根據所述意圖和所述需求方位置資訊對商戶進行初步篩選，獲得至少一個潛在商戶(步驟304)，並基於所述需求內容向潛在商戶的終端推送用來表示用戶端需求的音訊資料(步驟306)；
潛在商戶的終端播放所述音訊資料(步驟308)；
服務端基於潛在商戶的終端播放所述音訊資料後反饋的回應資訊，從篩選獲得的潛在商戶中篩選出滿足用戶端需求的目標商戶(步驟310)，並向用戶端推薦所述目標商戶的相關資訊(步驟312)。用戶端可以展示目標商戶的相關資訊。
其中，圖3A與圖2中相關技術相同，在此不一一贅述。
如圖3B所示，是本說明書根據一示例性實施例示出的一種離線即時需求處理的系統架構圖。在該示意圖中，用戶在用戶端透過文字/語音方式發送需求內容。服務端接收到攜帶需求內容的需求請求，如果需求內容是語音形式，透過語音辨識轉換成文字，並進行用戶意圖的識別。根據用戶意圖，以及商戶所屬預設類別，對商戶進行篩選，獲得潛在商戶。向潛在商戶發送用戶需求。如果用戶發送的是文字內容，則將文字內容透過語音合成而產生語音。商戶的離線設備接收到請求，進行播放。商戶根據自身情況，可以回應該請求。服務端接收到商戶的回應資訊，整理反饋資訊並發送至用戶端。如果是語音形式的回應，還可以判斷是否反饋該語音資料。用戶端接收到反饋資訊後，可以查看反饋資訊，並進行離線購買產品或服務。
由上述實施例可見，本說明書實施例提出了一個即時發送需求的移動功能。在服務端，透過語音辨識和意圖分析，準確獲取用戶的意圖。並提出了即時需求和周圍商戶進行匹配的演算法。該演算法能準確的找到能夠滿足該即時需求的周圍潛在商戶。特別是針對沒有有效手段招攬和營運周圍客戶的長尾小商戶，建構一個用戶端到商戶端的即使回應通道，解決雙方不能有效撮合、溝通的問題。
與前述資訊推薦方法的實施例相對應，本說明書還提供了資訊推薦裝置及其所應用的電子設備的實施例。
本說明書資訊推薦裝置的實施例可以應用在電腦設備，電腦設備可以為服務端設備。裝置實施例可以透過軟體來實現，也可以透過硬體或者軟硬體結合的方式來實現。以軟體實現為例，作為一個邏輯意義上的裝置，是透過其所在電腦設備的處理器將非易失性記憶體中對應的電腦程式指令讀取到記憶體中運行形成的。從硬體層面而言，如圖4所示，為本說明書資訊推薦裝置所在電腦設備的一種硬體結構圖，除了圖4所示的處理器410、網路介面420、記憶體430、以及非易失性記憶體440之外，實施例中資訊推薦裝置431所在的電腦設備通常根據該設備的實際功能，還可以包括其他硬體，對此不再贅述。
如圖5所示，是本說明書根據一示例性實施例示出的一種資訊推薦裝置的方塊圖，所述裝置包括：
意圖識別模組52，用以：對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，所述需求請求至少攜帶需求內容和需求方位置資訊；
初步篩選模組54，用以：至少根據所述意圖和所述需求方位置資訊對需求回應方進行初步篩選，獲得至少一個潛在需求回應方；
資訊傳輸模組56，用以：基於所述需求內容向潛在需求回應方推送需求提出方的需求；
目標篩選模組58，用以：基於潛在需求回應方反饋的回應資訊，從篩選獲得的潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方；
所述資訊傳輸模組56，還用以向需求提出方推薦所述目標需求回應方的相關資訊。
可選地，需求回應方和需求內容均按預設類別進行分類，所述意圖為所述需求內容對應的預設類別。
可選地，所述裝置還包括(圖5未示出)：
模型訓練模組，用以：從已知預設類別的需求回應方的資訊中，提取能表徵需求回應方所屬預設類別的特徵資料；根據已知預設類別的需求回應方所屬預設類別和特徵資料，建構得到類別預測模型；
類別預測模組，用以：透過類別預測模型預測未知預設類別的需求回應方所屬預設類別。
可選地，所述特徵資料至少包括預設時間段內收款頻率、預設時間段內收款金額分佈、預設時間段內收款時間分佈和所處地理位置中的一種或多種特徵。
可選地，需求提出方的需求至少包括：基於所述需求內容獲得的音訊資料，所述音訊資料在潛在需求回應方被即時播放。
可選地，所述目標需求回應方的相關資訊包括以下一種或多種資訊：
目標需求回應方的標識資訊、目標需求回應方反饋的答覆內容、目標需求回應方與需求提出方的距離資訊、需求提出方所處位置到達目標需求回應方所處位置的導航指引資料。
可選地，所述裝置還包括(圖5未示出)：
資訊標記模組，用於：將所述目標需求回應方的相關資訊，標記在地圖上與目標需求回應方的位置資訊相對應位置處；
所述資訊傳輸模組56用以向需求提出方發送已標記地圖。
對於裝置實施例而言，由於其基本對應於方法實施例，所以相關之處參見方法實施例的部分說明即可。以上所描述的裝置實施例僅僅是示意性的，其中所述作為分離部件說明的模組可以是或者也可以不是物理上分開的，作為模組顯示的部件可以是或者也可以不是物理模組，即可以位於一個地方，或者也可以分佈到多個網路模組上。可以根據實際的需要選擇其中的部分或者全部模組來實現本說明書方案的目的。本發明所屬技術領域中具有通常知識者在不付出創造性勞動的情況下，即可以理解並實施。
相應地，本說明書實施例還提供一種電腦設備，包括記憶體、處理器及儲存在記憶體上並可在處理器上運行的電腦程式，其中，所述處理器執行所述程式時實現如下方法：
對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，所述需求請求至少攜帶需求內容和需求方位置資訊；
至少根據所述意圖和所述需求方位置資訊對需求回應方進行初步篩選，獲得至少一個潛在需求回應方，並基於所述需求內容向潛在需求回應方推送需求提出方的需求；
基於潛在需求回應方反饋的回應資訊，從篩選獲得的潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方，並向需求提出方推薦所述目標需求回應方的相關資訊。
本說明書中的各個實施例均採用漸進的方式描述，各個實施例之間相同相似的部分互相參見即可，每個實施例重點說明的都是與其他實施例的不同之處。尤其，對於設備實施例而言，由於其基本相似於方法實施例，所以描述的比較簡單，相關之處參見方法實施例的部分說明即可。
一種電腦儲存媒體，所述儲存媒體中儲存有程式指令，所述程式指令包括：
對需求提出方發送的需求請求中的需求內容進行語義分析，獲得需求提出方的意圖，所述需求請求至少攜帶需求內容和需求方位置資訊；
至少根據所述意圖和所述需求方位置資訊對需求回應方進行初步篩選，獲得至少一個潛在需求回應方，並基於所述需求內容向潛在需求回應方推送需求提出方的需求；
基於潛在需求回應方反饋的回應資訊，從篩選獲得的潛在需求回應方中篩選出滿足需求提出方需求的目標需求回應方，並向需求提出方推薦所述目標需求回應方的相關資訊。
本說明書實施例可採用在一個或多個其中包含有程式碼的儲存媒體(包括但不限於磁碟記憶體、CD-ROM、光學記憶體等)上實施的電腦程式產品的形式。電腦可用儲存媒體包括永久性和非永久性、可移動和非可移動媒體，可以由任何方法或技術來實現資訊儲存。資訊可以是電腦可讀指令、資料結構、程式的模組或其他資料。電腦的儲存媒體的例子包括但不限於：相變記憶體(PRAM)、靜態隨機存取記憶體(SRAM)、動態隨機存取記憶體(DRAM)、其他類型的隨機存取記憶體(RAM)、唯讀記憶體(ROM)、電可擦除可編程唯讀記憶體(EEPROM)、快閃記憶體或其他記憶體技術、唯讀光碟唯讀記憶體(CD-ROM)、數位多功能光碟(DVD)或其他光學儲存、磁盒式磁帶，磁帶磁磁片儲存或其他磁性儲存裝置或任何其他非傳輸媒體，可用來儲存可以被計算設備存取的資訊。
本領域技術人員在考慮說明書及實踐這裡申請的發明後，將容易想到本說明書的其它實施方案。本說明書旨在涵蓋本說明書的任何變型、用途或者適應性變化，這些變型、用途或者適應性變化遵循本說明書的一般性原理並包括本說明書未申請的本技術領域中的公知常識或慣用技術手段。說明書和實施例僅被視為示例性的，本說明書的真正範圍和精神由下面的申請專利範圍所指出。
應當理解的是，本說明書並不局限於上面已經描述並在圖式中示出的精確結構，並且可以在不脫離其範圍進行各種修改和改變。本說明書的範圍僅由所附的申請專利範圍來限制。
以上所述僅為本說明書的較佳實施例而已，並不用來限制本說明書，凡在本說明書的精神和原則之內，所做的任何修改、等同替換、改進等，均應包含在本說明書保護的範圍之內。Exemplary embodiments will be described in detail here, examples of which are shown in the drawings. When the following description refers to drawings, unless otherwise indicated, the same numerals in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. Rather, they are merely examples of devices and methods that are consistent with some aspects of this specification as detailed in the scope of the attached patent applications.
The terminology used in this specification is for the purpose of describing particular embodiments only, and is not intended to limit this specification. The singular forms "a", "said" and "the" used in this specification and the appended patent applications are also intended to include most forms unless the context clearly indicates other meanings. It should also be understood that the term "and/or" as used herein refers to and includes any or all possible combinations of one or more associated listed items.
It should be understood that although the terms first, second, third, etc. may be used in this specification to describe various information, the information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this specification, the first information may also be referred to as second information, and similarly, the second information may also be referred to as first information. Depending on the context, the word "if" as used herein can be interpreted as "when" or "when..." or "responsive to certainty".
In practical applications, when a user puts forward a certain demand, it is often difficult to determine which demand response parties can respond to the demand. It is often necessary to conduct on-site inspections to confirm whether the demand response party can provide the service or product corresponding to the demand. It can be seen that the user obtains The efficiency of services or goods is low.
In view of this, the embodiments of the present specification provide an interactive channel between the demand requester and the demand responder to solve the problem that the two parties cannot effectively match and communicate, thereby improving the efficiency of users in obtaining services or products.
The technical solutions in the embodiments of the present specification will be exemplified and explained below with reference to the drawings in the embodiments of the present specification.
As shown in FIG. 1, it is a schematic diagram of an information recommendation system architecture according to an exemplary embodiment of this specification. In the schematic diagram, it may include a demand requesting end, a service end, and a demand response end. The demand requesting end may be a requesting end, for example, a user end. The demand requesting end may be an application program that can provide demand requesting services, such as Alipay and other applications. The demand requesting end may also be an electronic device with a demand requesting function. The electronic device can be a mobile phone or other handheld portable device, or a slightly smaller portable device such as a wristwatch device, pendant device, etc., or a miniaturized device, tablet computer, notebook computer, desktop computer , Computer or other electronic equipment integrated in computer monitor. The server can be a collective name for multiple server devices, or it can be a collective name for software installed on the server device. The demand response end can be the end that can respond to the corresponding demand, for example, it can be the merchant end. The demand response end can be an application program that can provide demand response services, or an electronic device with a demand response function.
Next, an example of the embodiments of this specification will be described from the perspective of the server. As shown in FIG. 2, it is a flowchart of an information recommendation method according to an exemplary embodiment of the present specification. The method includes:
In step 202, perform semantic analysis on the demand content in the demand request sent by the demand proposer to obtain the intention of the demand proposer. The demand request carries at least the demand content and the location information of the demand side;
In step 204, the demand responders are initially screened based on at least the intention and the demand party location information to obtain at least one potential demand responder, and based on the demand content, the demand proposer's demand is pushed to the potential demand responder ;
In step 206, based on the response information fed back by the potential demand responders, a target demand responder meeting the needs of the demand proposer is selected from the potential demand responders obtained by the screening, and the target demand responder is recommended to the demand proposer relevant information.
Among them, the demand proposer may be a user or a user terminal. The demand responder can be a merchant or a merchant. The demand content is used to describe the demand of the demand proposer, and it can be a voice content issued by the demand proposer or a text content. Demand-side location information can be the geographic location of the demand-side, or information used to determine the geographic location of the demand-side, such as WiFi information.
When performing semantic analysis on the demand content sent by the demand proposer, if the demand content is voice content, the demand content can be first recognized by speech to be parsed into text content, and the text content is semantically analyzed. The purpose of semantic analysis is to identify the intention of the requester. In one embodiment, the preset category corresponding to the demand content may be used as the intention of the demand proposer. For example, a preset category may be constructed in advance, and the preset category may also be called a category system or a preset type. Demand responders and demand content are classified according to preset categories. The preset category may be obtained based on the type of merchant, or obtained based on the user's intention. For example, the preset categories may include: department stores, food, and fruits.
In this embodiment, the preset category can be used as the intention of the demand proposer, that is, the purpose of performing semantic analysis on the demand content is to obtain the preset category corresponding to the demand content, so that the intention of the demand proposer can be quickly identified. By associating the demand content with the demand responder through the preset category, the demand responder can be preliminarily screened according to the preset category corresponding to the demand content.
There are many methods of semantic analysis. The following two examples are given as examples.
In one example, the intent of the demand proposer can be obtained by keyword matching. For example, performing semantic analysis on the demand content in the demand request sent by the demand proposer to obtain the intention of the demand proposer may include:
Extract keywords that match keywords in a preset correspondence relationship from the demand content, where the preset correspondence relationship is a correspondence relationship between a pre-constructed keyword and a preset category;
According to the preset correspondence, the preset category corresponding to the matched keyword is used as the intention of the demand proposer.
In this embodiment, the correspondence between keywords and preset categories may be constructed in advance. Correspondence can be obtained based on big data analysis. For example, it can be obtained according to the correspondence between the search term and the preset category to which the final product belongs in the history record of the user searching for the product on the e-commerce platform. In this embodiment, by using keyword matching, the preset category corresponding to the matched keyword in the demand content is used as the intention of the demand proposer, so that the intention of the demand proposer can be quickly obtained.
In another embodiment, the demand content may be input into a pre-built intent recognition model to obtain the preset category corresponding to the demand content. Among them, the intention recognition model may be a model constructed in advance to recognize the user's intention. For example, a training sample can be constructed using demand content of known preset categories, and a deep learning algorithm can be trained using the training sample to obtain an intent recognition model. In this embodiment, the intention recognition model is used to predict the preset category corresponding to the demand content, so as to realize the recognition of the preset category of more demand content.
Demand responders are classified according to preset categories. In practical applications, only some of the responders may obtain their preset categories, while some of the responders may not know their preset categories. For example, the preset categories to which some demand responders belong can be obtained based on the preset categories uploaded by the demand responders, or based on the attribute information of the demand responders in the e-commerce platform. For other demand responders who do not know the preset category clearly, they can use the demand responders who know the preset category to predict the category of the demand responders who do not know the preset category. For example, in one embodiment, the method further includes:
Extract the characteristic data that can represent the default category to which the demand responder belongs from the information of the demand responder of the known preset category;
According to the preset categories and characteristic data of the demand responders of known preset categories, construct a category prediction model;
Use the category prediction model to predict the default category to which the demand responders of unknown preset categories belong.
Among them, the information of the demand responder may include information related to the demand responder such as historical collection records, static attribute information and the like. The characteristic data may be characteristic data that can characterize a preset category to which the demand responder belongs. In one example, the characteristic data includes at least one or more characteristics of a collection frequency within a preset time period, a distribution of payment amount within a preset time period, a distribution time of payment within a preset time period, and a geographic location . It can be seen that the preset category of the demand responder can be reflected through the information such as the frequency of receipt, the amount of receipt, the time of receipt, and the location, so as to accurately predict the preset category of other demand responders. It should be understood that the characteristic data may also include other characteristic data, as long as it can characterize the preset category to which the demand responder belongs, which will not be repeated here.
According to the preset category and characteristic data of the demand responders of known preset categories, a category prediction model can be constructed. For example, the supervised algorithm is trained according to the preset category and characteristic data of the demand responder of the known preset category, and the category prediction model used to predict the preset category is obtained. Supervised algorithms can be linear algorithms, logistic regression, random forest, etc.
In this embodiment, through the above method, more preset categories of demand responders can be obtained, thereby broadening recommendable demand responders.
Further, in order to improve security, the demand responder referred to in the embodiments of this specification may be a responder satisfying the reliability condition. For example, the responders are screened based on historical information such as historical collection records to ensure that the demand responders obtained are reliable. In particular, demand responders are reliable merchants.
In the application phase, after obtaining the intention of the demand proposer, the demand responder may be initially screened according to the preset screening conditions to obtain at least one potential demand responder. Wherein, the filter factor of the preset filter condition includes at least the intention. In order to recommend the information related to the location of the demand originator, the filter factor of the preset filtering conditions also includes the location information of the demander. The initial screening of demand responders by intent may be to select the same demand responders as the preset category to which the intention belongs. Using the location information of the demand side to initially screen the demand responders may be to screen out the demand responders that are related to the distance from the demand proposer. For example, the demand responders within a preset range from the demand proposer are selected, or the demand responders belonging to the same area as the demand proposer are selected. It is understandable that other screening conditions can also be set to screen out potential demand responders suitable for recommendation to the demand proposer. For example, the preset filtering condition may further include: the potential demand responder is a demand responder whose terminal has an audio playback function, so that the terminal of the potential demand responder can instantly play the demand of the demand proposer. For example, the terminal of the potential demand responder may be an Alipay box or other devices of the same type with speaker functions.
In this embodiment, in order to distinguish different demand responders, the demand responders obtained through the preliminary screening are called potential demand responders. Through preliminary screening, the number of notification responders on the server side can be reduced, while disturbing unrelated demand responders.
After obtaining the potential demand responder, the demand demander can be pushed to the potential demand responder based on the content of the demand. In an example, the content of the demand can be directly pushed to the demand responder, so as to establish an interactive channel between the demand proposer and the demand responder, which is convenient for both parties to communicate. In another example, in order to increase the intensity of reminding the demand of the demand proposer, the demand proposer’s demand pushed to the demand responder at least includes the audio data obtained based on the demand content, so that the potential demand responder can play the audio data in real time. Facilitate immediate response. Specifically, if the demand content is voice data, the voice data is directly pushed to the potential demand responder; if the demand content is text data, the demand content is generated by voice synthesis to generate audio data, and the audio data is pushed to each potential Demand responder.
It can be seen that by promptly broadcasting the needs of the demand responder to the potential demand responder in the form of audio data, the potential responder can be reminded in time, so that the potential demand responder can make timely feedback and realize the real-time demand sent by the demand proposer Be promptly feedback.
After the potential demand responder’s equipment reminds the demand proposer’s demand, if the merchant meets the demand, the potential demand responder can respond to the demand. There are many types of responses. For example, whether the feedback can meet the demand requester's needs through physical buttons or key combinations; or, whether the content feedback through text or voice can meet the demand requester's needs.
After the server receives the response information fed back by the potential demand responder, it can filter out the target demand responders that meet the needs of the demand proposer from the potential demand responders, and recommend the relevant information of the target demand responder to the demand proposer .
Regarding the screening of target demand responders based on the response information, in one embodiment, the demand responders who can meet the needs of the demand proposer will only initiate the response information. Therefore, it is possible to determine whether the potential demand responder is based on whether the response information is received Meet the needs of the demander. Specifically, the potential demand responder who initiated the response information can be regarded as the target demand responder. If no response information is received from the potential demand responder, it may be preset that the potential demand responder does not meet the demand proposer's requirements. In another embodiment, the response information may include a response content of a potential responder's business feedback, for example, voice content or text content. On the one hand, according to the content of the reply, it can be analyzed whether the potential demand responder can meet the needs of the demand proposer, so that the potential demand responder who meets the demand can be regarded as the target demand responder. On the other hand, it can be preset that the potential demand responder that initiated the response information can meet the needs of the demand proposer. Therefore, the potential demand responder that initiated the response information is directly regarded as the target demand responder, and the reply content is regarded as the target demand response One of the relevant information of the party is pushed to the requester.
Recommend the relevant information of the target demand responder to the demand proposer, so that the users of the demand proposer can understand the target demand responder. In view of this, in one example, the relevant information of the target demand responder includes one or more of the following information:
The identification information of the target demand responder, the response content of the target demand responder feedback, the distance information between the target demand responder and the demand proposer, the navigation guidance data of the location of the demand proposer to reach the location of the target demand responder.
Among them, the identification information of the target demand responder may be an identification that uniquely identifies the merchant such as the merchant name. The location information of the target demand responder can be determined based on the GPS information, WiFi information, etc. in the response, or can be obtained by looking up the registered location information of the target demand responder based on the identification information.
It can be seen that by displaying the relevant information of the target demand responder, the user can be prevented from querying other information again, and the user's operation steps are reduced. In particular, when the relevant information includes the response content of the target demand responder, the communication between the demand proposer and the demand responder can be realized.
Further, recommending the target demand responder to the demand proposer includes: marking the relevant information of the target demand responder on the map at a position corresponding to the position information of the target demand responder, and sending it to the demand proposer Map marked.
It can be seen that by sending the marked map to the demand proposer, the relevant information of the target demand responder can be displayed in the demand proposer by using the map. If the merchant feeds back voice data, it can also display a control item that triggers voice playback at the corresponding position on the map to realize the interaction between the requester and the responder.
The various technical features in the above embodiments can be combined arbitrarily, as long as there is no conflict or contradiction between the combinations of the features, but the space is limited and the description is not carried out one by one. Therefore, any combination of various technical features in the above embodiments can also be combined arbitrarily. It belongs to the scope disclosed in this specification.
The following describes an example of the embodiments of this specification in conjunction with a specific application scenario.
As shown in FIG. 3A, it is a flowchart of an offline instantaneous demand processing method according to an exemplary embodiment of the present specification. The method includes:
The user terminal sends an offline demand request that carries at least demand content and demand-side location information to the server (step 302). The offline demand request may refer to a request that is an offline demand and needs to be processed immediately.
The server performs semantic analysis on the demand content to obtain the user's intention, at least preliminary screening merchants based on the intention and the demand side location information to obtain at least one potential merchant (step 304), and based on the demand content to The terminal of the potential merchant pushes the audio data used to represent the user's needs (step 306);
The terminal of the potential merchant plays the audio data (step 308);
Based on the response information fed back by the terminal of the potential merchant after playing the audio data, the server selects target merchants that meet the needs of the client from the potential merchants obtained by the screening (step 310), and recommends the relevant correlation of the target merchant to the client Information (step 312). The client can display relevant information about the target merchant.
Among them, FIG. 3A is the same as the related technology in FIG. 2 and will not be repeated here.
As shown in FIG. 3B, it is a system architecture diagram of off-line instant demand processing according to an exemplary embodiment of this specification. In this diagram, the user sends the required content through text/voice on the user side. The server receives the demand request carrying the demand content. If the demand content is in the form of voice, it is converted into text through voice recognition and the user's intention is recognized. According to the user's intention and the preset category to which the merchant belongs, the merchant is screened to obtain potential merchants. Send user needs to potential merchants. If the user sends text content, the text content is synthesized through speech to generate speech. The offline device of the merchant receives the request and plays it. Merchants can respond to requests according to their own situation. The server receives the merchant's response information, organizes the feedback information and sends it to the client. If the response is in the form of voice, you can also determine whether to feedback the voice data. After receiving feedback information, the client can view the feedback information and purchase products or services offline.
It can be seen from the above-mentioned embodiments that the embodiments of the present specification propose a mobile function for instant sending demand. On the server side, through voice recognition and intention analysis, the user's intention is accurately obtained. It also proposes an algorithm to match immediate needs with surrounding merchants. The algorithm can accurately find the surrounding potential merchants who can meet the immediate demand. Especially for long-tail small merchants who do not have effective means to solicit and operate customers around them, construct a response channel from the client to the client to solve the problem that the two parties cannot effectively match and communicate.
Corresponding to the foregoing embodiment of the information recommendation method, this specification also provides an embodiment of the information recommendation device and the electronic device applied thereto.
The embodiments of the information recommendation device in this specification can be applied to computer equipment, and the computer equipment can be server equipment. The device embodiments can be implemented by software, or by hardware or a combination of hardware and software. Taking software implementation as an example, as a logical device, it is formed by reading the corresponding computer program instructions in the non-volatile memory into the memory through the processor of the computer device where it is located and running. From the hardware level, as shown in FIG. 4, it is a hardware structure diagram of the computer equipment where the information recommendation device of this specification is located, except for the processor 410, network interface 420, memory 430, and non- In addition to the volatile memory 440, in the embodiment, the computer device where the information recommendation device 431 is located usually may include other hardware according to the actual function of the device, which will not be repeated here.
As shown in FIG. 5, it is a block diagram of an information recommendation device according to an exemplary embodiment of this specification. The device includes:
The intention identification module 52 is used to: perform semantic analysis on the demand content in the demand request sent by the demand proposer to obtain the intention of the demand proposer, and the demand request carries at least the demand content and the demand side location information;
The preliminary screening module 54 is used to: initially screen the demand responders according to the intention and the demand party position information to obtain at least one potential demand responder;
The information transmission module 56 is used to: push the demand requester's demand to potential demand responders based on the demand content;
The target screening module 58 is used to: based on the response information fed back by the potential demand responders, select the target demand responders satisfying the needs of the demand proposer from the potential demand responders obtained by the screening;
The information transmission module 56 is also used to recommend the relevant information of the target demand responder to the demand proposer.
Optionally, the demand responder and the demand content are classified according to preset categories, and the intention is the preset category corresponding to the demand content.
Optionally, the device further includes (not shown in FIG. 5):
The model training module is used to: extract the characteristic data that can represent the preset category of the demand responder from the information of the demand responder of the known preset category; the preset category of the demand responder according to the known preset category And feature data to construct a category prediction model;
The category prediction module is used to predict the default category to which the demand responder of the unknown preset category belongs through the category prediction model.
Optionally, the characteristic data includes at least one or more characteristics of a collection frequency within a preset time period, a distribution of payment amount within a preset time period, a distribution time of payment within a preset time period, and a geographical location.
Optionally, the demand of the demand proposer includes at least: audio data obtained based on the content of the demand, and the audio data is played in real time on the potential demand responder.
Optionally, the relevant information of the target demand responder includes one or more of the following information:
The identification information of the target demand responder, the response content of the target demand responder feedback, the distance information between the target demand responder and the demand proposer, the navigation guidance data of the location of the demand proposer to reach the location of the target demand responder.
Optionally, the device further includes (not shown in FIG. 5):
The information marking module is used to mark the relevant information of the target demand responder on the map at a position corresponding to the position information of the target demand responder;
The information transmission module 56 is used to send the marked map to the requester.
As for the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the description of the method embodiments. The device embodiments described above are only schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules. It can be located in one place, or it can be distributed on multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution in this specification. Those with ordinary knowledge in the technical field to which the present invention belongs can understand and implement without paying creative work.
Correspondingly, the embodiments of the present specification also provide a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the program to implement the following method :
Perform semantic analysis on the demand content in the demand request sent by the demand proposer to obtain the intention of the demand proposer. The demand request carries at least the demand content and the demand side location information;
At least preliminary screening the demand responders based on the intention and the demand party location information, obtaining at least one potential demand responder, and pushing the demand proposer's demand to the potential demand responder based on the demand content;
Based on the response information fed back by the potential demand responders, a target demand responder that meets the needs of the demand proposer is selected from the potential demand responders obtained by the screening, and the relevant information of the target demand responder is recommended to the demand proposer.
The embodiments in this specification are described in a gradual manner. The same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant part can be referred to the description of the method embodiment.
A computer storage medium, program instructions are stored in the storage medium, the program instructions include:
Perform semantic analysis on the demand content in the demand request sent by the demand proposer to obtain the intention of the demand proposer. The demand request carries at least the demand content and the demand side location information;
At least preliminary screening the demand responders based on the intention and the demand party location information, obtaining at least one potential demand responder, and pushing the demand proposer's demand to the potential demand responder based on the demand content;
Based on the response information fed back by the potential demand responders, a target demand responder that meets the needs of the demand proposer is selected from the potential demand responders obtained by the screening, and the relevant information of the target demand responder is recommended to the demand proposer.
Embodiments of this specification may take the form of computer program products implemented on one or more storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing program code. Available storage media for computers include permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. The information can be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM) , Read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic tape storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices.
Those skilled in the art will easily think of other embodiments of the specification after considering the specification and practicing the invention filed here. This specification is intended to cover any variations, uses, or adaptations of this specification. These variations, uses, or adaptations follow the general principles of this specification and include common general knowledge or common technical means in the technical field not applied in this specification. . The description and examples are only to be regarded as exemplary, and the true scope and spirit of this description are indicated by the following patent applications.
It should be understood that this specification is not limited to the precise structure that has been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of this specification is limited only by the scope of the attached patent application.
The above are only the preferred embodiments of this specification and are not intended to limit this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this specification should be included in this specification Within the scope of protection.
