本說明書一個或多個實施例提供一種基於地理位置的關鍵詞提取方法及裝置，用以實現關鍵詞抽取的多樣性。 為了使本技術領域的人員更好地理解本說明書一個或多個實施例中的技術方案，下面將結合本說明書一個或多個實施例中的附圖，對本說明書一個或多個實施例中的技術方案進行清楚、完整地描述，顯然，所描述的實施例僅僅是本說明書一部分實施例，而不是全部的實施例。基於本說明書一個或多個實施例，本領域普通技術人員在沒有作出進步性勞動前提下所獲得的所有其他實施例，都應當屬於本說明書一個或多個實施例保護的範圍。 圖1是根據本說明書一實施例的一種基於地理位置的關鍵詞提取方法的示意性流程圖，如圖1所示，該方法包括： 步驟S102，獲取目標使用者的地理位置資訊，根據地理位置資訊確定目標使用者的使用者類型。 其中，使用者類型包括位於指定地理位置的第一類使用者和/或位於非指定地理位置的第二類使用者。指定地理位置與非指定地理位置的劃分可基於是否跨境、是否位於某個國家、是否位於某個城市等。例如，當指定地理位置為國外(即跨境)時，非指定地理位置則為國內；當指定地理位置為北京時，非指定地理位置則為除北京之外的其他城市。 步驟S104，獲取第一類使用者基於指定地理位置的第一行為特徵詞，和/或，獲取第二類使用者基於非指定地理位置的第二行為特徵詞。 在一個實施例中，行為特徵詞可包括搜索詞、瀏覽詞等。其中，搜索詞包括目標使用者搜索的與指定地理位置相關的詞語，例如，在目標使用者針對指定地理位置(如大連)的搜索行為資訊中，記錄有目標使用者通過搜索“大連旅遊攻略”來查看相關搜索結果，則“大連旅遊攻略”即為搜索詞。瀏覽詞包括目標使用者瀏覽的文檔內容中的關鍵詞，例如，目標使用者瀏覽文檔“大連旅遊攻略”，那麼該文檔中的關鍵詞“旅遊攻略”、“濱海路”等即為瀏覽詞。 步驟S106，根據第一行為特徵詞和/或第二行為特徵詞，確定與指定地理位置相關的關鍵詞。 採用本說明書一個或多個實施例的技術方案，通過獲取目標使用者的地理位置資訊，根據地理位置資訊確定目標使用者的使用者類型，並獲取第一類使用者基於指定地理位置的第一行為特徵詞，和/或，獲取第二類使用者基於非指定地理位置的第二行為特徵詞，進而根據第一行為特徵詞和/或第二行為特徵詞確定與指定地理位置相關的關鍵詞。可見，該技術方案能夠基於各類目標使用者的地理位置資訊以及行為特徵詞自動挖掘出與指定地理位置相關的關鍵詞，相較於傳統的僅能從文檔中提取關鍵詞的方法而言，更加提高了關鍵詞提取的多樣性，且使得所提取的關鍵詞更能符合使用者行為，覆蓋率更廣。 以下詳細說明本發明實施例提供的一種基於地理位置的關鍵詞提取方法。 首先執行步驟S102，即獲取目標使用者的地理位置資訊，根據地理位置資訊確定目標使用者的使用者類型。 在一個實施例中，可根據LBS(Location Based Service，基於位置服務)獲取目標使用者的地理位置資訊。LBS是指通過電信移動運營商的無線電通訊網路或外部定位方式，獲取終端使用者的位置資訊，在GIS(Geographic Information System，地理資訊系統)平臺的支援下，為使用者提供相應服務的一種增值服務。 目標使用者的使用者類型包括位於指定地理位置的第一類使用者和/或位於非指定地理位置的第二類使用者。指定地理位置與非指定地理位置的劃分可基於是否跨境、是否位於某個國家、是否位於某個城市等。例如，當指定地理位置為國外(即跨境)時，非指定地理位置則為國內；當指定地理位置為北京時，非指定地理位置則為除北京之外的其他城市。 此外，這裡所說的“位於指定地理位置”並非是指使用者當前正位於指定地理位置，而是指獲取到的目標使用者的地理位置資訊中包含指定地理位置的資訊，即，目標使用者在某個時間段曾位於指定地理位置。若目標使用者在某個時間段曾位於指定地理位置，則可認為該目標使用者屬於第一類使用者；否則，若目標使用者不曾位於指定地理位置，則可認為該目標使用者屬於位於非指定地理位置的第二類使用者。因此，在獲取目標使用者的地理位置資訊時，可以僅獲取一段時間內目標使用者的地理位置資訊，如獲取最近一個月內目標使用者的地理位置資訊。 例如，指定地理位置為新加坡，若根據目標使用者在最近一個月內的地理位置資訊確定出目標使用者曾位於新加坡，則說明該目標使用者屬於位於指定地理位置——新加坡的第一類使用者；若根據目標使用者在最近一個月內的地理位置資訊確定出目標使用者不曾位於新加坡，則說明該目標使用者屬於位於非指定地理位置(即除新加坡之外的其他國家)的第二類使用者。 確定目標使用者的使用者類型之後，繼續執行步驟S104，即獲取第一類使用者基於指定地理位置的第一行為特徵詞，和/或，獲取第二類使用者基於非指定地理位置的第二行為特徵詞。 在一個實施例中，確定第一行為特徵詞時，可首先確定第一類使用者位於指定地理位置的時間；其次獲取第一類使用者在該時間之前的預設時間段內針對指定地理位置的行為資訊；然後根據該行為資訊確定第一行為特徵詞。 其中，行為資訊可包括搜索行為資訊、瀏覽行為資訊等。例如，指定地理位置為大連，則根據目標使用者執行的與大連相關的搜索行為(如搜索大連旅遊攻略)及瀏覽行為(如瀏覽大連美食推薦的文章)，可獲得目標使用者針對大連的搜索行為資訊及瀏覽行為資訊。 為使獲取到的行為資訊能確保關鍵詞提取的準確度，預設時間段通常不應設置為較長的時間段，可設置為目標使用者在位於指定地理位置的時間之前的10天、半個月或1個月內的時間段。例如，預設時間段為目標使用者在位於指定地理位置的時間之前的10天內的時間段，假設根據目標使用者的地理位置資訊，可確定出目標使用者位於指定地理位置——北京的時間為2017年10月15日，那麼可獲取2017年10月15日之前的10天內(即2017年10月5日至2017年10月15日之間)的時間段內針對指定地理位置——北京的行為資訊。 第一行為特徵詞包括搜索詞、瀏覽詞中的至少一項。其中，搜索詞包括第一類使用者搜索的與指定地理位置相關的詞語，例如，在第一類使用者針對指定地理位置(如大連)的搜索行為資訊中，記錄有第一類使用者通過搜索“大連旅遊攻略”來查看相關搜索結果，則“大連旅遊攻略”即為搜索詞。瀏覽詞包括第一類使用者瀏覽的文檔內容中的關鍵詞，例如，第一類使用者瀏覽文檔“大連旅遊攻略”，那麼該文檔中的關鍵詞“旅遊攻略”、“濱海路”等即為瀏覽詞。 第二行為特徵詞的獲取方式與第一行為特徵詞的獲取方式類似，不同之處在於，第二行為特徵詞針對的地理位置可能範圍較大。例如，若根據第二類使用者的地理位置資訊，獲知第二類使用者曾位於除指定地理位置——大連之外的北京、上海、三亞等城市，那麼可從第二類使用者搜索的與北京、上海、三亞等城市相關的詞語中抽取第二行為特徵詞，也可從第二類使用者所瀏覽的與北京、上海、三亞等城市相關的文檔內容中提取關鍵詞作為第二行為特徵詞。 獲取到第一行為特徵詞與第二行為特徵詞之後，繼續執行步驟S106，即根據第一行為特徵詞和/或第二行為特徵詞，確定與指定地理位置相關的關鍵詞。 在一個實施例中，確定與指定地理位置相關的關鍵詞時，可首先利用指定二分類演算法分別對多個第一行為特徵詞以及多個第二行為特徵詞進行訓練，以得到各第一行為特徵詞分別對指定地理位置的貢獻值；進而根據各第一行為特徵詞分別對指定地理位置的貢獻值，從多個第一行為特徵詞中選擇至少一個第一行為特徵詞作為與指定地理位置相關的關鍵詞。 其中，指定二分類演算法可包括邏輯回歸演算法、疊代決策樹演算法等任一種二分類演算法。具體的，可將多個第一行為特徵詞以及多個第二行為特徵詞作為指定二分類演算法的輸入，以使分類器針對輸入的資料進行訓練。由於邏輯回歸演算法、疊代決策樹演算法等二分類演算法為現有技術，因此不再贅述。 對多個第一行為特徵詞以及多個第二行為特徵詞進行訓練時，可根據各第一行為特徵詞分別在多個第一行為特徵詞以及在多個第二行為特徵詞中的出現率，確定各第一行為特徵詞分別對指定地理位置的貢獻值。其中，第一行為特徵詞的貢獻值與該第一行為特徵詞在多個第一行為特徵詞中的出現率成正比，與該第一行為特徵詞在多個第二行為特徵詞中的出現率成反比。 假設根據第一類使用者針對指定地理位置的行為資訊獲取到N個第一行為特徵詞，同時根據第二類使用者針對非指定地理位置的行為資訊獲取到M個第二行為特徵詞。若其中某個第一行為特徵詞在這N個第一行為特徵詞中的出現率較高，並且該第一行為特徵詞在M個第二行為特徵詞中的出現率較少，則說明該第一行為特徵詞對指定地理位置的貢獻值較高。 在確定各第一行為特徵詞對指定地理位置的貢獻值時，可預先設定在N個第一行為特徵詞中的出現率閾值X，以及在M個第二行為特徵詞中的出現率閾值Y。若某個第一行為特徵詞在N個第一行為特徵詞中的出現率高於出現率閾值X、且在M個第二行為特徵詞中的出現率低於出現率閾值Y，則可確定該第一行為特徵詞對指定地理位置的貢獻值高，進而可確定該第一行為特徵詞為與指定地理位置相關的關鍵詞。 第一行為特徵詞對指定地理位置的貢獻值可採用權重的方式進行表徵。權重範圍為0~1。即，在0~1範圍內，權重值越高，說明第一行為特徵詞對指定地理位置的貢獻值越高；反之，權重值越低，說明第一行為特徵詞對指定地理位置的貢獻值越低。權重值的高低與第一行為特徵詞在N個第一行為特徵詞中的出現率以及在M個第二行為特徵詞中的出現率有關，其中，第一行為特徵詞在N個第一行為特徵詞中的出現率越高、且在M個第二行為特徵詞中的出現率越低，則其對應的權重值越高。 需要說明的是，若某個第一行為特徵詞僅在N個第一行為特徵詞中的出現率高，而在M個第二行為特徵詞中的出現率不明確，則無法確定該第一行為特徵詞對指定地理位置的貢獻值。原因在於，某些行為特徵詞屬於指定地理位置與非指定地理位置可能共有的行為特徵詞，目標使用者無法是否位於指定地理位置，均有可能執行與該行為特徵詞相關的行為資訊。 例如，指定地理位置為國內，非指定地理位置為國外。假設獲取到的第一行為特徵詞和第二行為特徵詞中均包括“星巴克”一詞，由於目標使用者無論是在國內還是在國外均會執行與“星巴克”相關的行為資訊，例如搜索“星巴克”進行定位，或瀏覽與“星巴克”相關文章，因此行為特徵詞“星巴克”在第一行為特徵詞中的出現率和在第二行為特徵詞中的出現率均較高，這種情況下，行為特徵詞“星巴克”並不屬於與指定地理位置相關的關鍵詞。 以下採用兩個具體實施例來說明本說明書提供的基於地理位置的關鍵詞提取方法。 實施例一 圖2是根據本說明書具體實施例一的一種基於地理位置的關鍵詞提取方法的示意性流程圖。在該實施例一中，指定地理位置為國外，非指定地理位置為國內。如圖2所示，該方法包括： 步驟S201，獲取多個目標使用者在最近一段時間內的地理位置資訊，並根據地理位置資訊確定各目標使用者的使用者類型。 其中，使用者類型包括位於國外的第一類使用者(即跨境遊的使用者)和位於國內的第二類使用者。 步驟S202，獲取第一類使用者基於國外的N個第一行為特徵詞，以及，獲取第二類使用者基於國內的M個第二行為特徵詞。 該步驟中，獲取第一行為特徵詞時，可首先確定第一類使用者位於國外的時間，然後獲取第一類使用者在該時間之前的預設時間段內(例如在該時間之前的10天內)針對國外的行為資訊，如搜索或瀏覽與國外相關的關鍵詞、文章等，進而根據該行為資訊獲取第一行為特徵詞，例如將第一類使用者搜索的與國外相關的關鍵詞作為第一行為特徵詞，以及，將第一類使用者瀏覽的與國外相關的文章中的關鍵詞提取出來作為第一行為特徵詞。 獲取第二行為特徵詞時，可首先確定第二類使用者位於國內的時間。由於第二類使用者在最近一段時間內不曾位於國外，因此該時間即為地理位置資訊對應的時間；然後獲取第一類使用者在該時間之前的預設時間段內(例如在該時間之前的10天內)針對國內的行為資訊，如搜索或瀏覽與國內相關的關鍵詞、文章等，進而根據該行為資訊獲取第二行為特徵詞，例如將第二類使用者搜索的與國內相關的關鍵詞作為第二行為特徵詞，以及，將第二類使用者瀏覽的與國內相關的文章中的關鍵詞提取出來作為第二行為特徵詞。 步驟S203，利用指定二分類演算法分別對N個第一行為特徵詞以及M個第二行為特徵詞進行訓練。 步驟S204，在訓練過程中，判斷各第一行為特徵詞是否在N個第一行為特徵詞中的出現率高於預設閾值X、且在M個第二行為特徵詞中的出現率低於預設閾值Y；若是，則執行步驟S205；若否，則執行步驟S206。 步驟S205，確定該第一行為特徵詞為與跨境遊相關的關鍵詞。 步驟S206，確定該第一行為特徵詞不為與跨境遊相關的關鍵詞。 採用本說明書具體實施例一的技術方案，能夠基於第一類使用者基於國外的第一行為特徵詞、以及第二類使用者基於國內的第二行為特徵詞自動挖掘出與跨境遊相關的關鍵詞，相較於傳統的僅能從文檔中提取關鍵詞的方法而言，更加提高了關鍵詞提取的多樣性，且使得所提取的關鍵詞更能符合使用者行為，覆蓋率更廣。 實施例二 圖3是根據本說明書具體實施例二的一種基於地理位置的關鍵詞提取方法的示意性流程圖。在該實施例二中，指定地理位置為新加坡，非指定地理位置為除新加坡之外的其他國家。如圖3所示，該方法包括： 步驟S301，獲取多個目標使用者在最近一段時間內的地理位置資訊，並根據地理位置資訊確定各目標使用者的使用者類型。 其中，使用者類型包括位於新加坡的第一類使用者和位於其他國家的第二類使用者。 步驟S302，獲取第一類使用者基於新加坡的N個第一行為特徵詞，以及，獲取第二類使用者基於其他國家的M個第二行為特徵詞。 該步驟中，獲取第一行為特徵詞時，可首先確定第一類使用者位於新加坡的時間，然後獲取第一類使用者在該時間之前的預設時間段內(例如在該時間之前的10天內)針對新加坡的行為資訊，如搜索或瀏覽與新加坡相關的關鍵詞、文章等，進而根據該行為資訊獲取第一行為特徵詞，例如將第一類使用者搜索的與新加坡相關的關鍵詞作為第一行為特徵詞，以及，將第一類使用者瀏覽的與新加坡相關的文章中的關鍵詞提取出來作為第一行為特徵詞。 獲取第二行為特徵詞時，可首先確定第二類使用者位於其他國家的時間。由於第二類使用者在最近一段時間內不曾位於其他國家，因此該時間即為地理位置資訊對應的時間；然後獲取第一類使用者在該時間之前的預設時間段內(例如在該時間之前的10天內)針對其他國家的行為資訊，如搜索或瀏覽與其他國家相關的關鍵詞、文章等，進而根據該行為資訊獲取第二行為特徵詞，例如將第二類使用者搜索的與其他國家相關的關鍵詞作為第二行為特徵詞，以及，將第二類使用者瀏覽的與其他國家相關的文章中的關鍵詞提取出來作為第二行為特徵詞。 步驟S303，利用指定二分類演算法分別對N個第一行為特徵詞以及M個第二行為特徵詞進行訓練。 步驟S304，在訓練過程中，判斷各第一行為特徵詞是否在N個第一行為特徵詞中的出現率高於預設閾值X、且並未出現於M個第二行為特徵詞中；若是，則執行步驟S305；若否，則執行步驟S306。 本實施例二中，考慮到使用者去某個具體國家時的行為資訊更具有針對性，因此判斷第一行為特徵詞是否屬於與新加坡相關的關鍵詞時，判斷依據與跨境遊關鍵詞的判斷依據有所不同，即，只要某個第一行為特徵詞在N個第一行為特徵詞中的出現率較高、且並未出現於M個第二行為特徵詞中中，即可認為該第一行為特徵詞屬於與新加坡相關的關鍵詞。 當然，除上述判斷依據外，也可設定其他的判斷依據；例如，判斷各第一行為特徵詞是否在N個第一行為特徵詞中的出現率高於預設閾值X、且在M個第二行為特徵詞中的出現率低於預設閾值Y；或者，僅判斷各第一行為特徵詞是否在N個第一行為特徵詞中的出現率高於預設閾值X；等等。 步驟S305，確定該第一行為特徵詞為與新加坡相關的關鍵詞。 步驟S306，確定該第一行為特徵詞不為與新加坡相關的關鍵詞。 採用本說明書具體實施例二的技術方案，能夠基於第一類使用者基於新加坡的第一行為特徵詞、以及第二類使用者基於其他國家的第二行為特徵詞自動挖掘出與新加坡相關的關鍵詞，相較於傳統的僅能從文檔中提取關鍵詞的方法而言，更加提高了關鍵詞提取的多樣性，且使得所提取的關鍵詞更具有針對性、更能符合使用者行為，覆蓋率更廣。 綜上，已經對本主題的特定實施例進行了描述。其它實施例在所附申請專利範圍的範圍內。在一些情況下，在申請專利範圍中記載的動作可以按照不同的順序來執行並且仍然可以實現期望的結果。另外，在附圖中描繪的過程不一定要求示出的特定順序或者連續順序，以實現期望的結果。在某些實施方式中，多工處理和並行處理可以是有利的。 以上為本說明書一個或多個實施例提供的基於地理位置的關鍵詞提取方法，基於同樣的思路，本說明書一個或多個實施例還提供一種基於地理位置的關鍵詞提取裝置。 圖4是根據本說明書一實施例的一種基於地理位置的關鍵詞提取裝置的示意性方塊圖。如圖4所示，該裝置包括： 第一獲取模組410，獲取目標使用者的地理位置資訊，根據地理位置資訊確定目標使用者的使用者類型，其中，使用者類型包括位於指定地理位置的第一類使用者和/或位於非指定地理位置的第二類使用者； 第二獲取模組420，獲取第一類使用者基於指定地理位置的第一行為特徵詞，和/或，獲取第二類使用者基於非指定地理位置的第二行為特徵詞； 確定模組430，根據第一行為特徵詞和/或第二行為特徵詞，確定與指定地理位置相關的關鍵詞。 可選地，第二獲取模組420包括： 第一確定單元，確定第一類使用者位於指定地理位置的時間； 第一獲取單元，獲取第一類使用者在位於指定地理位置的時間之前的預設時間段內針對指定地理位置的行為資訊； 第二確定單元，根據行為資訊確定第一行為特徵詞。 可選地，行為資訊包括搜索行為資訊、瀏覽行為資訊中的至少一項；第一行為特徵詞包括搜索詞、瀏覽詞中的至少一項。 可選地，確定模組430包括： 訓練單元，利用指定二分類演算法分別對多個第一行為特徵詞以及多個第二行為特徵詞進行訓練，以得到各第一行為特徵詞分別對指定地理位置的貢獻值； 選擇單元，根據貢獻值，選擇至少一個第一行為特徵詞作為與指定地理位置相關的關鍵詞。 可選地，訓練單元還用於： 根據各第一行為特徵詞分別在多個第一行為特徵詞以及在多個第二行為特徵詞中的出現率，確定各第一行為特徵詞分別對指定地理位置的貢獻值； 其中，貢獻值與第一行為特徵詞在多個第一行為特徵詞中的出現率成正比，與第一行為特徵詞在多個第二行為特徵詞中的出現率成反比。 可選地，指定二分類演算法包括邏輯回歸演算法、疊代決策樹演算法中的至少一項。 可選地，指定地理位置為國外，非指定地理位置為國內。 可選地，指定地理位置為指定國家，非指定地理位置為除指定國家之外的其他國家。 可選地，第一獲取模組包括： 第二獲取單元，根據基於位置服務LBS獲取目標使用者的地理位置資訊。 採用本說明書一個或多個實施例的裝置，通過獲取目標使用者的地理位置資訊，根據地理位置資訊確定目標使用者的使用者類型，並獲取第一類使用者基於指定地理位置的第一行為特徵詞，和/或，獲取第二類使用者基於非指定地理位置的第二行為特徵詞，進而根據第一行為特徵詞和/或第二行為特徵詞確定與指定地理位置相關的關鍵詞。可見，該技術方案能夠基於各類目標使用者的地理位置資訊以及行為特徵詞自動挖掘出與指定地理位置相關的關鍵詞，相較於傳統的僅能從文檔中提取關鍵詞的方法而言，更加提高了關鍵詞提取的多樣性，且使得所提取的關鍵詞更能符合使用者行為，覆蓋率更廣。 本領域的技術人員應可理解，圖4中的基於地理位置的關鍵詞提取裝置能夠用來實現前文所述的基於地理位置的關鍵詞提取方法，其中的細節描述應與前文方法部分描述類似，為避免繁瑣，此處不另贅述。 基於同樣的思路，本說明書一個或多個實施例還提供一種基於地理位置的關鍵詞提取設備，如圖5所示。基於地理位置的關鍵詞提取設備可因配置或性能不同而產生比較大的差異，可以包括一個或一個以上的處理器501和記憶體502，記憶體502中可以儲存有一個或一個以上儲存應用程式或資料。其中，記憶體502可以是短暫儲存或持久儲存。儲存在記憶體502的應用程式可以包括一個或一個以上模組(圖示未示出)，每個模組可以包括對基於地理位置的關鍵詞提取設備中的一系列電腦可執行指令。更進一步地，處理器501可以設置為與記憶體502通信，在基於地理位置的關鍵詞提取設備上執行記憶體502中的一系列電腦可執行指令。基於地理位置的關鍵詞提取設備還可以包括一個或一個以上電源503，一個或一個以上有線或無線網路介面504，一個或一個以上輸入輸出介面505，一個或一個以上鍵盤506。 具體在本實施例中，基於地理位置的關鍵詞提取設備包括有記憶體，以及一個或一個以上的程式，其中一個或者一個以上程式儲存於記憶體中，且一個或者一個以上程式可以包括一個或一個以上模組，且每個模組可以包括對基於地理位置的關鍵詞提取設備中的一系列電腦可執行指令，且經配置以由一個或者一個以上處理器執行該一個或者一個以上套裝程式含用於進行以下電腦可執行指令： 獲取目標使用者的地理位置資訊，根據所述地理位置資訊確定所述目標使用者的使用者類型，其中，所述使用者類型包括位於指定地理位置的第一類使用者和/或位於非指定地理位置的第二類使用者； 獲取所述第一類使用者基於所述指定地理位置的第一行為特徵詞，和/或，獲取所述第二類使用者基於所述非指定地理位置的第二行為特徵詞； 根據所述第一行為特徵詞和/或所述第二行為特徵詞，確定與所述指定地理位置相關的關鍵詞。 可選地，電腦可執行指令在被執行時，還可以使所述處理器： 確定所述第一類使用者位於所述指定地理位置的時間； 獲取所述第一類使用者在位於所述指定地理位置的時間之前的預設時間段內針對所述指定地理位置的行為資訊； 根據所述行為資訊確定所述第一行為特徵詞。 可選地，所述行為資訊包括搜索行為資訊、瀏覽行為資訊中的至少一項；所述第一行為特徵詞包括搜索詞、瀏覽詞中的至少一項。 可選地，電腦可執行指令在被執行時，還可以使所述處理器： 利用指定二分類演算法分別對多個所述第一行為特徵詞以及多個所述第二行為特徵詞進行訓練，以得到各所述第一行為特徵詞分別對所述指定地理位置的貢獻值； 根據所述貢獻值，選擇至少一個所述第一行為特徵詞作為與所述指定地理位置相關的關鍵詞。 可選地，電腦可執行指令在被執行時，還可以使所述處理器： 根據各所述第一行為特徵詞分別在多個所述第一行為特徵詞以及在多個所述第二行為特徵詞中的出現率，確定各所述第一行為特徵詞分別對所述指定地理位置的貢獻值； 其中，所述貢獻值與所述第一行為特徵詞在多個所述第一行為特徵詞中的出現率成正比，與所述第一行為特徵詞在多個所述第二行為特徵詞中的出現率成反比。 可選地，所述指定二分類演算法包括邏輯回歸演算法、疊代決策樹演算法中的至少一項。 可選地，所述指定地理位置為國外，所述非指定地理位置為國內。 可選地，所述指定地理位置為指定國家，所述非指定地理位置為除所述指定國家之外的其他國家。 可選地，電腦可執行指令在被執行時，還可以使所述處理器： 根據基於位置服務LBS獲取所述目標使用者的地理位置資訊。 本說明書一個或多個實施例還提出了一種電腦可讀儲存媒體，該電腦可讀儲存媒體儲存一個或多個程式，該一個或多個程式包括指令，該指令當被包括多個應用程式的電子設備執行時，能夠使該電子設備執行上述基於地理位置的關鍵詞提取設備方法，並具體用於執行： 獲取目標使用者的地理位置資訊，根據所述地理位置資訊確定所述目標使用者的使用者類型，其中，所述使用者類型包括位於指定地理位置的第一類使用者和/或位於非指定地理位置的第二類使用者； 獲取所述第一類使用者基於所述指定地理位置的第一行為特徵詞，和/或，獲取所述第二類使用者基於所述非指定地理位置的第二行為特徵詞； 根據所述第一行為特徵詞和/或所述第二行為特徵詞，確定與所述指定地理位置相關的關鍵詞。 上述實施例闡明的系統、裝置、模組或單元，具體可以由電腦晶片或實體實現，或者由具有某種功能的產品來實現。一種典型的實現設備為電腦。具體的，電腦例如可以為個人電腦、膝上型電腦、蜂巢式電話、相機電話、智慧型電話、個人數位助理、媒體播放機、導航設備、電子郵件設備、遊戲控制台、平板電腦、可穿戴設備或者這些設備中的任何設備的組合。 為了描述的方便，描述以上裝置時以功能分為各種單元分別描述。當然，在實施本說明書一個或多個實施例時可以把各單元的功能在同一個或多個軟體和/或硬體中實現。 本領域內的技術人員應明白，本說明書一個或多個實施例可提供為方法、系統、或電腦程式產品。因此，本說明書一個或多個實施例可採用完全硬體實施例、完全軟體實施例、或結合軟體和硬體方面的實施例的形式。而且，本說明書一個或多個實施例可採用在一個或多個其中包含有電腦可用程式碼的電腦可用儲存媒體(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。 本說明書一個或多個實施例是參照根據本申請案實施例的方法、設備(系統)、和電腦程式產品的流程圖和／或方塊圖來描述的。應理解可由電腦程式指令實現流程圖和／或方塊圖中的每一流程和／或方塊、以及流程圖和／或方塊圖中的流程和／或方塊的結合。可提供這些電腦程式指令到通用電腦、專用電腦、嵌入式處理機或其他可程式設計資料處理設備的處理器以產生一個機器，使得通過電腦或其他可程式設計資料處理設備的處理器執行的指令產生用於實現在流程圖一個流程或多個流程和／或方塊圖一個方塊或多個方塊中指定的功能的裝置。 這些電腦程式指令也可儲存在能引導電腦或其他可程式設計資料處理設備以特定方式工作的電腦可讀記憶體中，使得儲存在該電腦可讀記憶體中的指令產生包括指令裝置的製造品，該指令裝置實現在流程圖一個流程或多個流程和／或方塊圖一個方塊或多個方塊中指定的功能。 這些電腦程式指令也可裝載到電腦或其他可程式設計資料處理設備上，使得在電腦或其他可程式設計設備上執行一系列操作步驟以產生電腦實現的處理，從而在電腦或其他可程式設計設備上執行的指令提供用於實現在流程圖一個流程或多個流程和／或方塊圖一個方塊或多個方塊中指定的功能的步驟。 在一個典型的配置中，計算設備包括一個或多個處理器(CPU)、輸入/輸出介面、網路介面和記憶體。 記憶體可能包括電腦可讀媒體中的非永久性記憶體，隨機存取記憶體(RAM)和/或非易失性記憶體等形式，如唯讀記憶體(ROM)或快閃記憶體(flash RAM)。記憶體是電腦可讀媒體的示例。 電腦可讀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)、電可擦除可程式設計唯讀記憶體(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，磁帶磁片儲存或其他磁性存放裝置或任何其他非傳輸媒體，可用於儲存可以被計算設備存取的資訊。按照本文中的界定，電腦可讀媒體不包括暫存電腦可讀媒體(transitory media)，如調變的資料信號和載波。 還需要說明的是，術語“包括”、“包含”或者其任何其他變體意在涵蓋非排他性的包含，從而使得包括一系列要素的過程、方法、商品或者設備不僅包括那些要素，而且還包括沒有明確列出的其他要素，或者是還包括為這種過程、方法、商品或者設備所固有的要素。在沒有更多限制的情況下，由語句“包括一個……”限定的要素，並不排除在包括所述要素的過程、方法、商品或者設備中還存在另外的相同要素。 本說明書一個或多個實施例可以在由電腦執行的電腦可執行指令的一般上下文中描述，例如程式模組。一般地，程式模組包括執行特定任務或實現特定抽象資料類型的常式、程式、物件、組件、資料結構等等。也可以在分散式運算環境中實踐本申請案，在這些分散式運算環境中，由通過通信網路而被連接的遠端處理設備來執行任務。在分散式運算環境中，程式模組可以位於包括存放裝置在內的本地和遠端電腦儲存媒體中。 本說明書中的各個實施例均採用遞進的方式描述，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，對於系統實施例而言，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。 以上所述僅為本說明書一個或多個實施例而已，並不用於限制本說明書。對於本領域技術人員來說，本說明書一個或多個實施例可以有各種更改和變化。凡在本說明書一個或多個實施例的精神和原理之內所作的任何修改、等同替換、改進等，均應包含在本說明書一個或多個實施例的申請專利範圍之內。One or more embodiments of this specification provide a method and device for extracting keywords based on geographic location, so as to realize the diversity of keyword extraction. In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will combine the drawings in one or more embodiments of this specification to compare The technical solution is described clearly and completely. Obviously, the described embodiments are only a part of the embodiments in this specification, rather than all the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by a person of ordinary skill in the art without making progressive work shall fall within the protection scope of one or more embodiments of this specification. Fig. 1 is a schematic flowchart of a method for extracting keywords based on geographic location according to an embodiment of the present specification. As shown in Fig. 1, the method includes: Step S102, obtaining geographic location information of a target user, according to geographic location The information determines the user type of the target user. Among them, the user type includes a first type of user located in a designated geographic location and/or a second type of user located in a non-designated geographic location. The division of designated geographic location and non-designated geographic location may be based on whether it is cross-border, whether it is located in a certain country, whether it is located in a certain city, etc. For example, when the designated geographic location is foreign (ie, cross-border), the non-designated geographic location is domestic; when the designated geographic location is Beijing, the non-designated geographic location is a city other than Beijing. Step S104: Acquire first behavior feature words of users of the first type based on the specified geographic location, and/or acquire second behavior feature words of users of the second category based on non-specified geographic locations. In an embodiment, the behavior characteristic words may include search words, browse words, and the like. Among them, the search terms include words related to the specified geographic location searched by the target user. For example, in the search behavior information of the target user for the specified geographic location (such as Dalian), it is recorded that the target user searches for "Dalian Tourism Strategy" To view related search results, "Dalian Travel Guide" is the search term. The browse words include keywords in the content of the document browsed by the target user. For example, if the target user browses the document "Dalian Travel Strategy", then the keywords "Travel Strategy" and "Binhai Road" in the document are the browse words. Step S106, according to the first behavior characteristic word and/or the second behavior characteristic word, a keyword related to the specified geographic location is determined. Using the technical solutions of one or more embodiments of this specification, by obtaining the geographic location information of the target user, the user type of the target user is determined based on the geographic location information, and the first type of user based on the specified geographic location is obtained. Behavior feature words, and/or, obtain second behavior feature words based on non-specified geographic locations of users of the second category, and then determine keywords related to the specified geographic location based on the first behavior feature words and/or second behavior feature words . It can be seen that this technical solution can automatically dig out keywords related to a specified geographic location based on the geographic location information and behavioral characteristic words of various target users. Compared with the traditional method that only extracts keywords from a document, The diversity of keyword extraction is further improved, and the extracted keywords are more in line with user behaviors, and the coverage rate is wider. The following describes in detail a method for extracting keywords based on geographic location provided by an embodiment of the present invention. First, step S102 is performed, which is to obtain geographic location information of the target user, and determine the user type of the target user according to the geographic location information. In one embodiment, the geographic location information of the target user can be obtained according to LBS (Location Based Service). LBS refers to obtaining the location information of end users through the radio communication network or external positioning method of the telecommunications mobile operator, and with the support of the GIS (Geographic Information System) platform, a value-added service that provides users with corresponding services service. The user types of target users include the first type of users located in a designated geographic location and/or the second type of users located in a non-designated geographic location. The division of designated geographic location and non-designated geographic location may be based on whether it is cross-border, whether it is located in a certain country, whether it is located in a certain city, etc. For example, when the designated geographic location is foreign (ie, cross-border), the non-designated geographic location is domestic; when the designated geographic location is Beijing, the non-designated geographic location is a city other than Beijing. In addition, the "located in a specified geographic location" mentioned here does not mean that the user is currently located in the specified geographic location, but refers to the location information of the target user obtained that contains the information of the specified geographic location, that is, the target user Was in a specified geographic location during a certain period of time. If the target user has been located in the specified geographic location in a certain period of time, the target user can be considered to belong to the first category of users; otherwise, if the target user has not been located in the specified geographic location, the target user can be considered to be located in The second category of users who are not geographically specified. Therefore, when obtaining the geographic location information of the target user, it is possible to obtain only the geographic location information of the target user within a period of time, such as obtaining the geographic location information of the target user in the most recent month. For example, if the designated geographic location is Singapore, if it is determined that the target user was located in Singapore based on the geographic location information of the target user in the most recent month, it means that the target user belongs to the designated geographic location—the first type of use in Singapore If it is determined that the target user has never been located in Singapore based on the geographic location information of the target user in the most recent month, it means that the target user belongs to the second place located in a non-designated geographic location (that is, a country other than Singapore) Class user. After determining the user type of the target user, proceed to step S104, that is, obtain the first behavior feature words of the first type of users based on the specified geographic location, and/or obtain the second type of users based on the non-specified geographic location. 2. Behavior characteristic words. In one embodiment, when determining the first behavior feature word, first determine the time when the first type of user is located in the designated geographic location; secondly, obtain the first type of user's specific geographic location within a preset time period before that time The behavior information; and then determine the first behavior characteristic word based on the behavior information. Among them, the behavior information may include search behavior information, browsing behavior information, etc. For example, if the designated geographic location is Dalian, then based on the Dalian-related search behaviors (such as searching for Dalian travel guides) and browsing behaviors (such as browsing articles recommended by Dalian cuisine) performed by the target user, the target user's search for Dalian can be obtained Behavior information and browsing behavior information. In order for the acquired behavior information to ensure the accuracy of keyword extraction, the preset time period should generally not be set to a longer period of time. It can be set to 10 days or a half before the time when the target user is located in the specified geographic location. A period of time within a month or a month. For example, the preset time period is the time period within 10 days before the time when the target user is located in the specified geographic location. Assuming that the target user is located in the specified geographic location-Beijing based on the geographic location information of the target user The time is October 15, 2017, then you can get the specified geographic location within 10 days before October 15, 2017 (that is, between October 5, 2017 and October 15, 2017)— —Beijing behavior information. The first behavior characteristic word includes at least one of a search word and a browse word. Among them, the search terms include words related to the specified geographic location searched by the first category of users. For example, in the search behavior information of the first category of users for the specified geographic location (such as Dalian), it is recorded that the first category of users passed Search for "Dalian Travel Strategy" to view related search results, then "Dalian Travel Strategy" is the search term. Browsing words include keywords in the content of documents browsed by users of the first category. For example, if users of the first category browse the document "Dalian Travel Strategy", then the keywords in the document "Travel Strategy", "Binhai Road", etc. For browsing words. The acquisition method of the second behavior characteristic word is similar to the acquisition method of the first behavior characteristic word, except that the geographic location targeted by the second behavior characteristic word may be larger. For example, if based on the geographic location information of the second type of user, it is known that the second type of user was located in Beijing, Shanghai, Sanya and other cities other than the designated geographic location-Dalian, then you can search from the second type of user The second behavior feature words are extracted from words related to Beijing, Shanghai, Sanya and other cities, and keywords can be extracted as the second behavior from the content of documents related to Beijing, Shanghai, Sanya and other cities browsed by users of the second type. Feature words. After the first behavior characteristic word and the second behavior characteristic word are obtained, step S106 is continued, that is, based on the first behavior characteristic word and/or the second behavior characteristic word, keywords related to the designated geographic location are determined. In one embodiment, when determining keywords related to a specified geographic location, a specified two-class classification algorithm may be used to train multiple first behavior feature words and multiple second behavior feature words to obtain each first The contribution value of the behavior feature words to the specified geographic location respectively; and then according to the contribution value of each first behavior feature word to the specified geographic location, at least one first behavior feature word is selected from the multiple first behavior feature words as the Location-related keywords. Among them, the designated two-classification algorithm may include any two-classification algorithm such as logistic regression algorithm and iterative decision tree algorithm. Specifically, multiple first behavior feature words and multiple second behavior feature words can be used as the input of the specified two-class classification algorithm, so that the classifier is trained on the input data. Since binary classification algorithms such as logistic regression algorithm and iterative decision tree algorithm are existing technologies, they will not be described in detail. When training multiple first behavior feature words and multiple second behavior feature words, the occurrence rate of each first behavior feature word in multiple first behavior feature words and multiple second behavior feature words can be used according to , To determine the contribution value of each first behavior feature word to the specified geographic location. Among them, the contribution value of the first behavior characteristic word is proportional to the appearance rate of the first behavior characteristic word in multiple first behavior characteristic words, and is proportional to the appearance of the first behavior characteristic word in multiple second behavior characteristic words The rate is inversely proportional. Suppose that N first behavior characteristic words are obtained according to the behavior information of the first type of users for the specified geographic location, and M second behavior characteristic words are obtained according to the behavior information of the second type of users for the non-specified geographic location. If one of the first behavior feature words has a higher occurrence rate in the N first behavior feature words, and the first behavior feature word has a lower occurrence rate in the M second behavior feature words, it means that The first behavior feature word has a higher contribution value to the specified geographic location. When determining the contribution value of each first behavior feature word to the specified geographic location, the occurrence rate threshold X in N first behavior feature words and the occurrence rate threshold Y in M second behavior feature words can be preset . If the occurrence rate of a certain first behavior characteristic word in N first behavior characteristic words is higher than the occurrence rate threshold X, and the occurrence rate of M second behavior characteristic words is lower than the occurrence rate threshold Y, it can be determined The first behavior characteristic word has a high contribution value to the specified geographic location, and it can be determined that the first behavior characteristic word is a keyword related to the specified geographic location. The contribution value of the first behavior feature word to the specified geographic location can be characterized by weight. The weight range is 0~1. That is, in the range of 0~1, the higher the weight value, the higher the contribution value of the first behavior feature word to the specified geographic location; otherwise, the lower the weight value, the higher the contribution value of the first behavior feature word to the specified geographic location. The lower. The weight value is related to the occurrence rate of the first behavior feature word in N first behavior feature words and the occurrence rate of M second behavior feature words, where the first behavior feature word is in the N first behavior words The higher the occurrence rate of feature words and the lower the occurrence rate of M second behavior feature words, the higher the corresponding weight value. It should be noted that if a certain first behavior characteristic word only has a high occurrence rate in N first behavior characteristic words, but the occurrence rate of M second behavior characteristic words is not clear, then the first behavior characteristic word cannot be determined. The contribution value of the behavior feature words to the specified geographic location. The reason is that some behavior characteristic words belong to behavior characteristic words that may be shared between a designated geographic location and a non-designated geographic location, and the target user cannot be located in the designated geographic location, and it is possible to implement behavior information related to the behavior characteristic word. For example, the designated geographic location is domestic, and the non-designated geographic location is foreign. Assuming that the acquired first behavior characteristic words and second behavior characteristic words both include the word "Starbucks", since the target users will perform "Starbucks"-related behavior information, such as search "Starbucks" locates or browses articles related to "Starbucks". Therefore, the occurrence rate of the behavior characteristic word "Starbucks" in the first behavior characteristic word and the second behavior characteristic word are both higher. In this case Below, the behavioral characteristic word "Starbucks" does not belong to keywords related to the specified geographic location. The following two specific embodiments are used to illustrate the geographic location-based keyword extraction method provided in this specification. Embodiment 1 Fig. 2 is a schematic flowchart of a method for extracting keywords based on geographic location according to Embodiment 1 of this specification. In the first embodiment, the designated geographic location is abroad, and the non-designated geographic location is domestic. As shown in FIG. 2, the method includes: Step S201: Obtain geographic location information of multiple target users in a recent period of time, and determine the user type of each target user based on the geographic location information. Among them, the types of users include the first type of users located abroad (ie, cross-border travel users) and the second type of users located in China. Step S202, acquiring N first behavior characteristic words of users of the first type based on foreign countries, and acquiring M second behavior characteristic words of users of the second type based on China. In this step, when acquiring the first behavior feature words, you can first determine the time when the first type of user is located abroad, and then acquire the first type of user within a preset time period before that time (for example, the time before that time). Within 10 days) For foreign behavior information, such as searching or browsing foreign-related keywords, articles, etc., and then obtaining the first behavior feature words based on the behavior information, for example, searching for the first category of users with foreign Related keywords are used as the first behavior feature words, and keywords in articles related to foreign countries browsed by the first type of users are extracted as the first behavior feature words. When acquiring the second behavior feature words, the time when the second type of user is located in the country can be determined first. Since users of the second type have not been located abroad in the most recent period of time, this time is the time corresponding to the geographic location information; then the user of the first type is acquired within the preset time period before that time (for example, at that time In the previous 10 days) for domestic behavior information, such as searching or browsing domestic-related keywords, articles, etc., and then obtaining second behavior characteristic words based on the behavior information, for example, searching for the second category of users related to domestic As the second behavior feature word, and extract keywords from domestic-related articles browsed by the second category of users as the second behavior feature word. In step S203, the N first behavior feature words and M second behavior feature words are trained separately by using a specified two-class classification algorithm. Step S204: In the training process, it is determined whether the occurrence rate of each first behavior characteristic word in the N first behavior characteristic words is higher than a preset threshold X, and the occurrence rate of each first behavior characteristic word in the M second behavior characteristic words is lower than The preset threshold Y; if yes, go to step S205; if no, go to step S206. Step S205: Determine that the first behavior characteristic word is a keyword related to cross-border travel. Step S206: Determine that the first behavior feature word is not a keyword related to cross-border travel. Using the technical solution of specific embodiment 1 of this specification, it is possible to automatically dig out related cross-border travel based on the first type of user’s first behavior feature words based on foreign countries and the second type of user’s based on domestic second behavior feature words Compared with the traditional method that only extracts keywords from documents, the keywords of, increase the diversity of keyword extraction, and make the extracted keywords more in line with user behaviors and wider coverage . Embodiment 2 Fig. 3 is a schematic flowchart of a method for extracting keywords based on geographic location according to Embodiment 2 of this specification. In the second embodiment, the designated geographic location is Singapore, and the non-designated geographic location is countries other than Singapore. As shown in FIG. 3, the method includes: Step S301: Obtain geographic location information of multiple target users in a recent period of time, and determine the user type of each target user based on the geographic location information. Among them, the user types include the first type of users located in Singapore and the second type of users located in other countries. Step S302: Obtain N first behavior characteristic words of users of the first type based on Singapore, and obtain M second behavior characteristic words of users of the second type based on other countries. In this step, when acquiring the first behavior feature words, first determine the time when the first type of user is located in Singapore, and then acquire the first type of user within a preset time period before that time (for example, 10 years before that time). Within days) behavioral information for Singapore, such as searching or browsing keywords and articles related to Singapore, and then obtaining the first behavioral characteristic words based on the behavioral information, for example, searching for keywords related to Singapore from the first category of users As the first behavior feature word, and extract keywords from articles related to Singapore browsed by the first category of users as the first behavior feature word. When acquiring the second behavior feature words, the time when the second type of users are located in other countries can be determined first. Since users of the second type have not been located in other countries in the most recent period of time, the time is the time corresponding to the geographic location information; then the user of the first type is acquired within the preset time period before that time (for example, at that time In the previous 10 days) for the behavior information of other countries, such as searching or browsing keywords and articles related to other countries, and then obtaining the second behavior characteristic words based on the behavior information, for example, the second type of user search and Keywords related to other countries are used as second behavior feature words, and keywords in articles related to other countries browsed by users of the second category are extracted as second behavior feature words. In step S303, the N first behavior feature words and M second behavior feature words are trained separately by using a designated two-class classification algorithm. Step S304, in the training process, determine whether the occurrence rate of each first behavior characteristic word in the N first behavior characteristic words is higher than the preset threshold X and does not appear in the M second behavior characteristic words; if so; , Go to step S305; if not, go to step S306. In the second embodiment, considering that the user’s behavior information when going to a specific country is more targeted, when determining whether the first behavior feature word belongs to a keyword related to Singapore, the judgment basis is based on the cross-border travel keyword. The judgment basis is different, that is, as long as the occurrence rate of a certain first behavior characteristic word in N first behavior characteristic words is high, and it does not appear in M second behavior characteristic words, it can be considered The first behavior feature words are keywords related to Singapore. Of course, in addition to the above judgment basis, other judgment basis can also be set; for example, it is judged whether the occurrence rate of each first behavior characteristic word in the N first behavior characteristic words is higher than the preset threshold X, and is in the M th The occurrence rate of the second behavior feature words is lower than the preset threshold Y; or, it is only judged whether the occurrence rate of each first behavior feature word in the N first behavior feature words is higher than the preset threshold X; etc. Step S305: Determine that the first behavior feature word is a keyword related to Singapore. Step S306: It is determined that the first behavior characteristic word is not a keyword related to Singapore. Using the technical solution of specific embodiment 2 of this specification, it is possible to automatically dig out the key related to Singapore based on the first behavior feature words of the first type of users based on Singapore and the second type of users based on the second behavior feature words of other countries Compared with the traditional method that only extracts keywords from documents, it improves the diversity of keyword extraction, and makes the extracted keywords more targeted, more in line with user behavior, and covering The rate is wider. In summary, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the attached patent application. In some cases, the actions described in the scope of the patent application can be performed in a different order and still achieve desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown in order to achieve the desired result. In certain embodiments, multiplexing and parallel processing may be advantageous. The above is the geographic location-based keyword extraction method provided by one or more embodiments of this specification. Based on the same idea, one or more embodiments of this specification also provide a geographic location-based keyword extraction device. Fig. 4 is a schematic block diagram of an apparatus for extracting keywords based on geographic location according to an embodiment of the present specification. As shown in FIG. 4, the device includes: a first acquisition module 410, which acquires geographic location information of the target user, and determines the user type of the target user according to the geographic location information, wherein the user type includes those located in a specified geographic location The first type of user and/or the second type of user located in a non-designated geographic location; the second acquisition module 420, which obtains the first behavior feature words of the first type of user based on the specified geographic location, and/or, obtains the first type of user The second type of user is based on the second behavior feature words of the non-specified geographic location; the determining module 430 determines keywords related to the specified geographic location based on the first behavior feature words and/or the second behavior feature words. Optionally, the second acquisition module 420 includes: a first determining unit, which determines the time when a user of the first type is located at a designated geographic location; a first acquiring unit, which acquires the time before the time when a user of the first type is located at the designated geographic location Behavior information for the designated geographic location within a preset time period; the second determining unit determines the first behavior characteristic word according to the behavior information. Optionally, the behavior information includes at least one of search behavior information and browsing behavior information; and the first behavior characteristic word includes at least one of a search word and a browsing word. Optionally, the determining module 430 includes: a training unit, which uses a specified two-class classification algorithm to train a plurality of first behavior feature words and a plurality of second behavior feature words respectively, so as to obtain that each first behavior feature word is assigned to The contribution value of the geographic location; the selection unit, according to the contribution value, selects at least one first behavior characteristic word as a keyword related to the specified geographic location. Optionally, the training unit is further configured to: according to the respective occurrence rates of each first behavior feature word in multiple first behavior feature words and multiple second behavior feature words, determine that each first behavior feature word is assigned to The contribution value of the geographic location; where the contribution value is proportional to the occurrence rate of the first behavior feature word in multiple first behavior feature words, and is proportional to the occurrence rate of the first behavior feature word in multiple second behavior feature words Inversely. Optionally, the specified binary classification algorithm includes at least one of a logistic regression algorithm and an iterative decision tree algorithm. Optionally, the designated geographic location is foreign and the non-designated geographic location is domestic. Optionally, the designated geographic location is a designated country, and the non-designated geographic location is a country other than the designated country. Optionally, the first acquiring module includes: a second acquiring unit that acquires geographic location information of the target user according to the location-based service LBS. Using the device of one or more embodiments of this specification, by obtaining the geographic location information of the target user, the user type of the target user is determined according to the geographic location information, and the first behavior of the first type of user based on the specified geographic location is obtained Feature words, and/or, acquire second behavior feature words of users of the second category based on non-specified geographic locations, and then determine keywords related to the specified geographic location based on the first behavior feature words and/or second behavior feature words. It can be seen that this technical solution can automatically dig out keywords related to a specified geographic location based on the geographic location information and behavioral characteristic words of various target users. Compared with the traditional method that only extracts keywords from a document, The diversity of keyword extraction is further improved, and the extracted keywords are more in line with user behaviors, and the coverage rate is wider. Those skilled in the art should understand that the geographic location-based keyword extraction device in Figure 4 can be used to implement the geographic location-based keyword extraction method described above, and the detailed description should be similar to the previous method part description. To avoid cumbersomeness, I won’t repeat them here. Based on the same idea, one or more embodiments of this specification also provide a geographic location-based keyword extraction device, as shown in FIG. 5. Geographical location-based keyword extraction devices may have relatively large differences due to different configurations or performances, and may include one or more processors 501 and memory 502, and memory 502 may store one or more storage applications Or information. The memory 502 may be short-term storage or permanent storage. The application program stored in the memory 502 may include one or more modules (not shown in the figure), and each module may include a series of computer-executable instructions in the geographic location-based keyword extraction device. Furthermore, the processor 501 may be configured to communicate with the memory 502, and execute a series of computer-executable instructions in the memory 502 on a geographic location-based keyword extraction device. The geographic location-based keyword extraction device may also include one or more power sources 503, one or more wired or wireless network interfaces 504, one or more input and output interfaces 505, and one or more keyboards 506. Specifically, in this embodiment, the device for extracting keywords based on geographic location includes memory and one or more programs. One or more programs are stored in the memory, and one or more programs may include one or more programs. More than one module, and each module may include a series of computer-executable instructions in a geographic location-based keyword extraction device, and is configured to be executed by one or more processors to execute the one or more package programs. It is used to perform the following computer-executable commands: acquiring geographic location information of a target user, and determining the user type of the target user based on the geographic location information, wherein the user type includes the first location located at a specified geographic location Type users and/or second type users located in non-designated geographic locations; obtain the first behavior feature words of the first type users based on the specified geographic location, and/or obtain the second type use The second behavior feature word based on the non-designated geographic location; and the keyword related to the designated geographic location is determined according to the first behavior feature word and/or the second behavior feature word. Optionally, when the computer-executable instructions are executed, the processor may also cause the processor to: determine the time when the user of the first type is located in the designated geographic location; Behavior information for the designated geographic location within a preset time period before the time of the designated geographic location; and determine the first behavior characteristic word according to the behavior information. Optionally, the behavior information includes at least one of search behavior information and browsing behavior information; and the first behavior characteristic word includes at least one of search words and browsing words. Optionally, when the computer-executable instructions are executed, the processor can also be used to train a plurality of the first behavior characteristic words and a plurality of the second behavior characteristic words respectively by using a specified two-class classification algorithm , To obtain the contribution value of each of the first behavior characteristic words to the designated geographic location; according to the contribution value, select at least one of the first behavior characteristic words as a keyword related to the designated geographic location. Optionally, when the computer-executable instructions are executed, the processor may also cause the processor: according to each of the first behavior characteristic words, respectively, in a plurality of the first behavior characteristic words and in a plurality of the second behavior characteristic words The occurrence rate of the characteristic words is used to determine the contribution value of each of the first behavior characteristic words to the designated geographic location; wherein, the contribution value and the first behavior characteristic word are in a plurality of the first behavior characteristics The appearance rate in a word is proportional to the appearance rate of the first behavior characteristic word in the plurality of second behavior characteristic words. Optionally, the specified binary classification algorithm includes at least one of a logistic regression algorithm and an iterative decision tree algorithm. Optionally, the designated geographic location is abroad, and the non-designated geographic location is domestic. Optionally, the designated geographic location is a designated country, and the non-designated geographic location is a country other than the designated country. Optionally, when the computer-executable instructions are executed, the processor may also: obtain the geographic location information of the target user according to the location-based service LBS. One or more embodiments of this specification also propose a computer-readable storage medium that stores one or more programs, and the one or more programs include instructions. When the instructions are included in multiple application programs When the electronic device is executed, the electronic device can execute the above-mentioned geographic location-based keyword extraction device method, and is specifically used to execute: obtain geographic location information of a target user, and determine the target user's geographic location information based on the geographic location information User type, wherein the user type includes a first type of user located at a designated geographic location and/or a second type of user located at a non-designated geographic location; and obtaining the first type of user based on the specified geographic location The first behavior feature word of the location, and/or, obtain the second behavior feature word of the second type of user based on the non-designated geographic location; according to the first behavior feature word and/or the second behavior Feature words, to determine keywords related to the specified geographic location. The systems, devices, modules or units explained in the above embodiments may be implemented by computer chips or entities, or implemented by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, and a wearable. Device or any combination of these devices. For the convenience of description, when describing the above device, the functions are divided into various units and described separately. Of course, when implementing one or more embodiments of this specification, the functions of each unit may be implemented in the same or multiple software and/or hardware. Those skilled in the art should understand that one or more embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, one or more embodiments of this specification may adopt the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware. Moreover, one or more embodiments of this specification can be implemented on one or more computer-usable storage media (including but not limited to magnetic disk memory, CD-ROM, optical memory, etc.) containing computer-usable program codes. In the form of a computer program product. One or more embodiments of this specification are described with reference to flowcharts and/or block diagrams of methods, equipment (systems), and computer program products according to the embodiments of this application. It should be understood that each process and/or block in the flowchart and/or block diagram, and the combination of processes and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, dedicated computers, embedded processors, or other programmable data processing equipment to generate a machine that can be executed by the processor of the computer or other programmable data processing equipment Produce means for realizing the functions specified in one or more processes in the flowchart and/or one block or more in the block diagram. These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including the instruction device , The instruction device realizes the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram. These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer or other programmable equipment to generate computer-implemented processing, so that the computer or other programmable equipment The instructions executed above provide steps for implementing functions specified in one or more processes in the flowchart and/or one block or more in the block diagram. In a typical configuration, the computing device includes one or more processors (CPU), input/output interfaces, network interfaces, and memory. Memory may include non-permanent memory in computer-readable media, random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash memory ( flash RAM). Memory is an example of computer-readable media. Computer-readable media includes permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology. Information can be computer-readable instructions, data structures, program modules, 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), and 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 CD-ROM (CD-ROM), digital multi-function Optical discs (DVD) or other optical storage, magnetic cassettes, 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. According to the definition in this article, computer-readable media does not include transitory media, such as modulated data signals and carrier waves. It should also be noted that the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or equipment including a series of elements not only includes those elements, but also includes Other elements that are not explicitly listed, or they also include elements inherent to such processes, methods, commodities, or equipment. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, commodity, or equipment that includes the element. One or more embodiments of this specification may be described in the general context of computer-executable instructions executed by a computer, such as a program module. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. It is also possible to practice this application in a distributed computing environment. In these distributed computing environments, remote processing devices connected through a communication network perform tasks. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices. The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, as for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiment. The above description is only one or more embodiments of this specification, and is not used to limit this specification. For those skilled in the art, one or more embodiments of this specification can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of patent application of one or more embodiments of this specification.
