TWI688873B - Enviromental information question and answer system and method - Google Patents
Enviromental information question and answer system and method Download PDFInfo
- Publication number
- TWI688873B TWI688873B TW108100803A TW108100803A TWI688873B TW I688873 B TWI688873 B TW I688873B TW 108100803 A TW108100803 A TW 108100803A TW 108100803 A TW108100803 A TW 108100803A TW I688873 B TWI688873 B TW I688873B
- Authority
- TW
- Taiwan
- Prior art keywords
- data
- environmental
- auxiliary
- processor
- target
- Prior art date
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
本發明係關於一種環境資訊問答系統與方法,特別是一種基於人工智慧的環境資訊問答系統與方法。The invention relates to an environmental information question answering system and method, in particular to an environmental intelligence question answering system and method based on artificial intelligence.
隨著人工智慧與通訊軟體的發展,使用具有人工智慧的機器人在通訊軟體上,執行簡單應答或查詢資料等工作,已漸漸成為趨勢。使用者可透過通訊軟體或其他介面傳送訊息給機器人,機器人則分析訊息內容所包含的關鍵字,並根據關鍵字搜尋資料庫,以提供使用者所需的資訊。With the development of artificial intelligence and communication software, the use of artificial intelligence robots on communication software to perform simple responses or query data has gradually become a trend. Users can send messages to the robot through communication software or other interfaces. The robot analyzes the keywords contained in the message content and searches the database based on the keywords to provide the information the user needs.
然而,當上述的裝置應用在防災時,由於氣象相關的資料過於龐大,因此機器人在搜尋資料時需要耗費較多的時間或硬體資源;在此情形下,當使用者傳送的訊息不夠完整(例如,訊息中缺少時間、地點或明確的氣象資訊),機器人可能會耗費更多時間搜尋,且找出的結果極可能非使用者實際上欲詢求的資訊。此情況不僅造成使用者在使用過程的不便(使用者可能因對機器人的操作理解不足而未能在第一時間提供完整的訊息),也可能在使用者一直未能提供完整訊息的情況下導致使用者無法取得所需的正確氣象資訊,因而未能即時預測和評估自然災害,影響防災的效果。However, when the above-mentioned device is used in disaster prevention, because the meteorological data is too large, the robot needs to spend more time or hardware resources when searching for the data; in this case, when the message sent by the user is not complete ( For example, the lack of time, location or clear weather information in the message), the robot may spend more time searching, and the result found is most likely not the information that the user actually wants. This situation not only causes inconvenience to the user in the use process (the user may fail to provide complete information in the first place due to insufficient understanding of the operation of the robot), but may also cause the user to fail to provide complete information Users cannot obtain the correct meteorological information they need, so they cannot predict and evaluate natural disasters in real time, affecting the effectiveness of disaster prevention.
因此,目前尚需要一種環境資訊問答系統與方法,以改善上述問題。Therefore, there is still a need for an environmental information question answering system and method to improve the above problems.
本發明在於提供一種環境資訊問答系統與方法,當使用者傳送的訊息不夠完整,此系統與方法能依據不同的氣象狀況,對使用者的問題做回應,以減少機器人搜尋資料所耗費的時間和硬體資源。The present invention is to provide an environmental information question answering system and method. When the information sent by the user is not complete, the system and method can respond to the user's questions according to different weather conditions, so as to reduce the time and time spent by the robot in searching data Hardware resources.
本發明在於提供一種環境資訊問答方法,包含:以該訊息收發介面依據所接收的語言訊息產生並傳送一語言輸入指令到一處理器,其中該語言輸入指令包含一目標環境資料與一輔助環境資料的至少一者;以該處理器的一語意分析元件接收該語言輸入指令所包含的該目標環境資料與該輔助環境資料的該至少一者,並選擇性地以該處理器產生該語言輸入指令未包含的該目標環境資料或該輔助環境資料,且以該語意分析元件分析該目標環境資料與該輔助環境資料;以該處理器依據一環境知識庫的多筆資料點,分析該目標環境資料與該輔助環境資料之間的一最佳關聯性;以該處理器依據該最佳關聯性產生一查詢指令,並依據該查詢指令與該環境知識庫產生一回應指令;以該處理器傳送該回應指令到該訊息收發介面;以及以接收該回應指令的該訊息收發介面呈現該回應訊息。The invention provides a question and answer method for environmental information, including: generating and transmitting a language input command to a processor according to the received language message using the message sending and receiving interface, wherein the language input command includes a target environment data and an auxiliary environment data At least one of; receiving the at least one of the target environment data and the auxiliary environment data included in the language input command by a semantic analysis component of the processor, and selectively generating the language input command by the processor The target environmental data or the auxiliary environmental data that are not included, and analyze the target environmental data and the auxiliary environmental data with the semantic analysis element; the processor analyzes the target environmental data based on multiple data points of an environmental knowledge base An optimal relationship with the auxiliary environment data; the processor generates a query command based on the optimal relationship, and generates a response command based on the query command and the environmental knowledge base; the processor transmits the Responding to the message sending and receiving interface; and presenting the reply message with the message receiving and sending interface receiving the responding command.
本發明在於提供一種環境資訊問答方法,包含:一訊息收發介面,用於依據所接收的語言訊息產生一語言輸入指令以及接收並呈現一回應訊息,其中該語言輸入指令包含一目標環境資料與一輔助環境資料的至少一者;一處理器,與該訊息收發介面電性連結,並包含一語意分析元件,該處理器用於接收該語言輸入指令,並選擇性地產生該語言輸入指令未包含的該目標環境資料或該輔助環境資料,且以該語意分析元件分析該目標環境資料與該輔助環境資料;以及一環境知識庫,與該處理器互相連結,並包含多筆資料點;其中該處理器根據該環境知識庫的該些資料點,分析該目標環境資料與該輔助環境資料之間的一最佳關聯性, 並依據該最佳關聯性產生一查詢指令,並依據該查詢指令與該環境知識庫產生並傳送該回應訊息至該訊息收發介面。The invention provides a question and answer method for environmental information, including: a message sending and receiving interface for generating a language input command and receiving and presenting a response message according to the received language message, wherein the language input command includes a target environment data and a At least one of the auxiliary environmental data; a processor electrically connected to the message transceiving interface and including a semantic analysis component, the processor is used to receive the language input command and selectively generate the language input command not included The target environmental data or the auxiliary environmental data, and analyzing the target environmental data and the auxiliary environmental data with the semantic analysis element; and an environmental knowledge base, interconnected with the processor, and including multiple data points; wherein the processing The device analyzes an optimal correlation between the target environmental data and the auxiliary environmental data according to the data points of the environmental knowledge base, and generates a query command according to the optimal correlation, and according to the query command and the The environmental knowledge base generates and sends the response message to the message receiving and sending interface.
本發明在於提供一種環境資訊問答系統與方法,當使用者傳送的訊息不夠完整,此系統與方法能依據不同的氣象狀況,對使用者的問題做回應,有效地減少機器人搜尋資料所耗費的時間和硬體資源,讓使用者能更精確地根據不同的氣象狀況預測和評估自然災害。The present invention is to provide an environmental information question answering system and method. When the information transmitted by the user is not complete, the system and method can respond to the user's questions according to different weather conditions, effectively reducing the time spent by the robot searching for data And hardware resources, so that users can more accurately predict and evaluate natural disasters according to different weather conditions.
以上之關於本揭露內容之說明及以下之實施方式之說明係用以示範與解釋本發明之精神與原理,並且提供本發明之專利申請範圍更進一步之解釋。The above description of the disclosure and the following description of the embodiments are used to demonstrate and explain the spirit and principle of the present invention, and provide a further explanation of the scope of the patent application of the present invention.
以下在實施方式中詳細敘述本發明之詳細特徵以及優點,其內容足以使任何熟習相關技藝者了解本發明之技術內容並據以實施,且根據本說明書所揭露之內容、申請專利範圍及圖式,任何熟習相關技藝者可輕易地理解本發明相關之目的及優點。以下之實施例係進一步詳細說明本發明之觀點,但非以任何觀點限制本發明之範疇。The following describes in detail the detailed features and advantages of the present invention in the embodiments. The content is sufficient for any person skilled in the relevant art to understand and implement the technical content of the present invention, and according to the contents disclosed in this specification, the scope of patent application and the drawings Anyone skilled in the relevant art can easily understand the purpose and advantages of the present invention. The following examples further illustrate the views of the present invention in detail, but do not limit the scope of the present invention in any way.
請參考圖1,圖1為依據本發明一實施例的環境資訊問答系統1的結構圖。如圖1所示,環境資訊問答系統1,包含訊息收發介面11、處理器13以及環境知識庫15。訊息收發介面11可以是具有接收語言訊息功能的介面,訊息收發介面並可依據所接收的語言訊息產生語言輸入指令,以及在環境資訊問答系統1的運作過程中接收回應訊息,並將回應訊息以文字或語音的形式呈現給使用者。舉例來說,訊息收發介面11可以是能接收或產生文字訊息的通訊軟體或網頁,或是能收發語音訊息的聲電轉換組件(例如:麥克風和喇叭)。需注意的是,因本發明的環境資訊問答系統1係應用在環境查詢與後續的災害評估與防災決策,因此語言輸入指令可包含目標環境資料與輔助環境資料。具體來說,上述的目標環境資料可以是與水利、氣象、地質相關的資料,或其他與相關自然災害的法規條文,例如:全國抽水機分佈狀況、雨量、氣溫、地震規模或是其他與防災相關的法規條文,而上述的輔助環境資料則可以是查詢目標環境資料所需的時間點、地點等資訊。換言之,若使用者提供的資訊係屬完整,則語言輸入指令係包含目標環境資料與輔助環境資料;然而,在使用者提供的資訊不夠完整的情況下,語言輸入指令只包含目標環境資料與輔助環境資料的其中之一。Please refer to FIG. 1, which is a structural diagram of an environmental information
處理器13可以是中央處理器(central processing unit,CPU)或任一種具計算功能的元件,而此處理器13包含語意分析元件131,並且與訊息收發介面11電性連接。在環境資訊問答系統1的運作過程中,處理器13可接收來自訊息收發介面11的語言輸入指令,並當語言輸入指令未同時包含目標環境資料與輔助環境資料時,選擇性地產生語言輸入指令未包含的目標環境資料或輔助環境資料,並以語意分析元件131分析目標環境資料與輔助環境資料。The
環境知識庫15可以是任一種具儲存功能並與處理器13電性連接的實體硬碟,也可以是與處理器13通訊連結的雲端硬碟,並包含多筆資料點與多個資料路徑,其中前述的目標環境資料與輔助環境資料係關聯於該些資料點,且該些資料點係藉由該些資料路徑互相連接。在環境資訊問答系統1的運作過程中,環境知識庫15用於提供上述的資料點給處理器13,以使處理器13能依據該些資料點分析目標環境資料與輔助環境資料之間的最佳關聯性, 並依據最佳關聯性產生查詢指令,再根據查詢指令搜尋儲存於環境知識庫15的資料以產生回應訊息,並將回應訊息傳送至訊息收發介面11。The
承上所述,為能有效率地從環境知識庫15中找出使用者所需的資料,處理器13先以「最短路徑演算法」計算出目標環境資料與輔助環境資料之間的最佳關聯性,再將該最佳關聯性轉換為一系列的程式碼,做為前述的查詢指令。因此,處理器13可依據查詢指令,直接從環境知識庫15得到使用者所需的資訊。詳細來說,環境知識庫15包含多筆資料點與連接該些資料點的多個資料路徑,並形成一「知識模型」;處理器13則將儲存於環境知識庫15的資料點與資料路徑視為不同的節點與連線,並以使用者提供的輔助環境資料為起點,以使用者查詢的目標環境資料為終點,從上述的資料點中找出關聯性最強的兩資料點(一般可定義為資料路徑的總數或總和為最小值的一搜尋路徑),並將該路徑定義為上述的「最佳關聯性」。在找出最佳關聯性後,處理器13可將關聯於該最佳關聯性的該搜尋路徑,轉換為具體的程式指令(即為前述之「查詢指令」),並依據該轉換而得的程式指令,進入環境知識庫15找出該路徑所關聯的環境資料。As mentioned above, in order to efficiently find the data required by the user from the
為能具體地描述環境資訊問答系統1的運作流程,請參考圖2,圖2為依據本發明一實施例的環境資訊問答方法的流程圖。當使用者透過訊息收發介面11輸入包含語言訊息的文字指令或語音指令時,請參考步驟S11:訊息收發介面11依據所接收的語言訊息產生並傳送語言輸入指令到處理器13。當處理器13接收訊息收發介面11所傳送的語言輸入指令,請參考步驟S13:處理器13的語意分析元件131接收語言輸入指令所包含的目標環境資料與輔助環境資料的至少一者;其中目標環境資料為使用者欲詢問的氣象數據,例如雨量或溫度,輔助環境資料則為使用者欲詢問氣象數據的時間或地點。此外,語意分析元件131能分析輸入指令所包含的目標環境資料與輔助環境資料,以使處理器13能判斷語言輸入指令是否同時包含目標環境資料與輔助環境資料。當語言輸入指令不同時包含目標環境資料與輔助環境資料時,請參考步驟S15:處理器13選擇性地產生語言輸入指令未包含的目標環境資料或輔助環境資料,並以語意分析元件131分析目標環境資料與輔助環境資料。舉例來說,當使用者詢問「台北的雨量多少?」,環境資訊問答系統1先以語意分析元件131分析上述的語言輸入指令:目標環境資料為「雨量」且輔助環境資料為「台北」,處理器13再依此判斷該語言輸入指令尚缺少另一輔助環境資料「時間」,並自行產生一個時間資料(例如:一小時內),以補足語言輸入指令所缺少的資訊。To specifically describe the operation flow of the environmental information
承前段落所述,當語言輸入指令同時包含目標環境資料與輔助環境資料時,請接續S17:處理器13依據環境知識庫15的多筆資料點分析目標環境資料與輔助環境資料之間的最佳關聯性。簡單來說,當使用者詢問「台北的雨量多少?」,環境知識庫15裡可能包含與「台北」相關的多筆資料點,以及與「雨量」相關的多筆資料點,而處理器13可將搜尋所需時間最短的兩資料點視為最佳關聯性。接續,請參考步驟S19:處理器13依據最佳關聯性產生查詢指令,並依據查詢指令與環境知識庫15產生回應指令;其中查詢指令可由程式碼所組成,以便處理器13依據查詢指令從環境知識庫15中搜尋使用者所需的資料,並依據搜尋結果產生回應指令。於處理器13產生回應指令後,請參考步驟S21:處理器13傳送回應指令到訊息收發介面11,以及步驟S23:訊息收發介面11接收並呈現回應訊息,以供使用者閱讀或聆聽環境資訊問答系統1所提供的氣象資訊。As mentioned in the previous paragraph, when the language input command includes both the target environment data and the auxiliary environment data, please continue to S17: the
為更詳細地說明圖2的步驟S15與步驟S17,請參考圖3。圖3為依據本發明一實施例的環境資訊問答方法的細部流程圖。承前所述,語言輸入指令可能會因為使用者所提供的資訊不夠完整,而缺少目標環境資料或輔助環境資料。請參考步驟S151:當處理器13判斷語言輸入指令不包含輔助環境資料,處理器13以預設輔助條件或即時環境條件產生輔助環境資料;其中預設輔助條件係預先儲存於處理器13,即時環境條件則為選擇性地被儲存於環境知識庫15,當處理器13判斷環境知識庫15不具有即時環境條件,則以預設輔助條件產生輔助環境資料。具體來說,預設輔助條件可以是預設的時間(例如:最近的一小時內)或是預設的地點(例如:全台灣),當使用者僅詢問「雨量多少?」,處理器13可將其解讀為「全台灣在最近的一小時內雨量多少?」。另一方面,即時環境條件則可以是政府機關所發布的特殊氣象警報,例如豪雨警報。由於在特殊氣象警報期間,會有大量有防災決策需求的使用者查詢氣象狀況;因此,當使用者在豪雨警報時期間詢問「台北雨量多少?」,處理器13可自動提供三小時內的台北雨量,以便使用者根據有效的數據進行防災評估。此外,請參考步驟S153:當處理器13判斷語言輸入指令不包含目標環境資料,處理器13以預設氣象資料產生目標環境資料。簡單來說,目標環境資料可以是大眾詢問度最高的氣象數據;例如,當使用者詢問「今天台北天氣如何?」而使語言輸入指令缺少目標環境資料,處理器13可自行解讀為「今天台北的氣溫多少?」,並從環境知識庫15中搜尋相關資料。To explain step S15 and step S17 of FIG. 2 in more detail, please refer to FIG. 3. FIG. 3 is a detailed flowchart of an environmental information question answering method according to an embodiment of the invention. As mentioned above, language input commands may lack target environment data or auxiliary environment data due to incomplete information provided by users. Please refer to step S151: when the
請繼續參考圖3。當語言輸入指令已包含目標環境資料和輔助環境資料,請參考步驟S171:處理器13依據環境知識庫15的多筆資料點,產生目標環境資料與輔助環境資料之間的多個關聯性;其中該些資料點之間係以多個資料路徑互相連結,且目標環境資料與輔助環境資料係關聯於該些資料點。於處理器13產生目標環境資料與輔助環境資料之間的多個關聯性後,請接續步驟S173:處理器13計算該些關聯性的多個資料路徑的總和,並以總和為最小值的該關聯性做為最佳關聯性。簡單來說,當兩資料點之間的該些資料路徑的總和為最小值時,表示此兩筆資料點之間的關聯性最強,且處理器13搜尋資料所需的時間也最短;因此,藉由步驟S173,環境資訊問答系統1可有效率地提供使用者所需的氣象資料。Please continue to refer to Figure 3. When the language input command already includes the target environment data and the auxiliary environment data, please refer to step S171: the
此外,值得一提的是,因本環境資訊問答系統1提供查詢的資料並不限於氣象資料,凡舉與「自然環境」相關的資訊,例如:水利與災害相關法規條文、全國抽水機分佈狀況等資訊,皆能透過環境資訊問答系統1查訊,故前述的目標環境資料與輔助環境資料並不限於本揭露書中所舉例的氣象資料,且環境知識庫15所儲存的資料點除了能和氣象資料相關,更可以與水利與災害相關法規條文、全國抽水機分佈狀況或等資訊相關,本發明不以此為限。In addition, it is worth mentioning that the information provided by this environmental information question and
綜上所述,本發明在於提供一種環境資訊問答系統1與方法,當使用者傳送的訊息不夠完整,此系統與方法能依據不同的氣象狀況,對使用者的問題做回應,有效地減少機器人搜尋資料所耗費的時間和硬體資源,讓使用者能更精確地根據不同的氣象狀況預測和評估自然災害。In summary, the present invention is to provide an environmental information
雖然本發明以前述之實施例揭露如上,然其並非用以限定本發明。在不脫離本發明之精神和範圍內,所為之更動與潤飾,均屬本發明之專利保護範圍。關於本發明所界定之保護範圍請參考所附之申請專利範圍。Although the present invention is disclosed as the foregoing embodiments, it is not intended to limit the present invention. Without departing from the spirit and scope of the present invention, all modifications and retouching are within the scope of patent protection of the present invention. For the protection scope defined by the present invention, please refer to the attached patent application scope.
1:環境資訊問答系統 11:訊息收發介面 13:處理器 15:環境知識庫 131:語意分析元件 1: Environmental information question and answer system 11: messaging interface 13: processor 15: Environmental Knowledge Base 131: Semantic analysis components
圖1為依據本發明一實施例的環境資訊問答系統的結構圖。 圖2為依據本發明一實施例的環境資訊問答方法的流程圖。 圖3為依據本發明一實施例的環境資訊問答方法的細部流程圖。 FIG. 1 is a structural diagram of an environmental information question answering system according to an embodiment of the present invention. 2 is a flowchart of an environmental information question answering method according to an embodiment of the invention. FIG. 3 is a detailed flowchart of an environmental information question answering method according to an embodiment of the invention.
Claims (9)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108100803A TWI688873B (en) | 2019-01-09 | 2019-01-09 | Enviromental information question and answer system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108100803A TWI688873B (en) | 2019-01-09 | 2019-01-09 | Enviromental information question and answer system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI688873B true TWI688873B (en) | 2020-03-21 |
TW202026912A TW202026912A (en) | 2020-07-16 |
Family
ID=70766941
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW108100803A TWI688873B (en) | 2019-01-09 | 2019-01-09 | Enviromental information question and answer system and method |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI688873B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWM547715U (en) * | 2017-05-08 | 2017-08-21 | 日盛證券股份有限公司 | Online robot customer service system |
TWM552625U (en) * | 2017-06-23 | 2017-12-01 | Urad Co Ltd | Advertising system with smart recommendations |
TWM555527U (en) * | 2017-10-02 | 2018-02-11 | 好險網科技股份有限公司 | Smart agent robot system |
TW201833730A (en) * | 2017-03-02 | 2018-09-16 | 大陸商騰訊科技(深圳)有限公司 | Information interaction method and device |
-
2019
- 2019-01-09 TW TW108100803A patent/TWI688873B/en active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW201833730A (en) * | 2017-03-02 | 2018-09-16 | 大陸商騰訊科技(深圳)有限公司 | Information interaction method and device |
TWM547715U (en) * | 2017-05-08 | 2017-08-21 | 日盛證券股份有限公司 | Online robot customer service system |
TWM552625U (en) * | 2017-06-23 | 2017-12-01 | Urad Co Ltd | Advertising system with smart recommendations |
TWM555527U (en) * | 2017-10-02 | 2018-02-11 | 好險網科技股份有限公司 | Smart agent robot system |
Also Published As
Publication number | Publication date |
---|---|
TW202026912A (en) | 2020-07-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11138212B2 (en) | Natural language response recommendation clustering for rapid retrieval | |
US20210342549A1 (en) | Method for training semantic analysis model, electronic device and storage medium | |
JP2020521210A (en) | Information processing method and terminal, computer storage medium | |
CN110019736B (en) | Question-answer matching method, system, equipment and storage medium based on language model | |
US20210357587A1 (en) | An intelligent response method and device | |
WO2019223362A1 (en) | Automatic answering method and device | |
US20180082680A1 (en) | Syntactic re-ranking of potential transcriptions during automatic speech recognition | |
US10558931B2 (en) | Determining comprehensiveness of question paper given syllabus | |
US20220138770A1 (en) | Method and apparatus for analyzing sales conversation based on voice recognition | |
CN111368029B (en) | Interaction method, device and equipment based on intention triples and storage medium | |
CN110140140A (en) | The detection and commission of psychomotor task | |
TWI688873B (en) | Enviromental information question and answer system and method | |
CN117332068B (en) | Man-machine interaction method and device, electronic equipment and storage medium | |
US11762937B2 (en) | Information processing apparatus, information processing system, and method of processing information | |
WO2020119346A1 (en) | Natural semantic comprehension method and apparatus, and computing device | |
US20230222120A1 (en) | Methods and apparatus for retrieving information via an intermediate representation | |
US10585960B2 (en) | Predicting locations for web pages and related techniques | |
US10083173B2 (en) | Artificial intelligence based language interpretation system | |
Khan et al. | Making Personal Digital Assistants Aware of What They Do Not Know. | |
CN111209391A (en) | Information identification model establishing method and system and interception method and system | |
WO2020151318A1 (en) | Corpus construction method and apparatus based on crawler model, and computer device | |
JP7443649B2 (en) | Model update method, device, electronic device and storage medium | |
CN114818683B (en) | Operation and maintenance method and device based on mobile terminal | |
CN113344550B (en) | Flow processing method, device, equipment and storage medium | |
KR102694794B1 (en) | Question-and-answer system, server and operation method for generating answer to user question based on rag technique |