TW202006563A - Dialogic type search display method performing an interactive search and obtain a corresponding search result by means of natural speech or natural sentence expression - Google Patents

Abstract

This invention relates to a dialogic type search display method, which includes the following steps: (a) capturing an input content of a user; (b) analyzing the input content according to a natural language algorithm and analyzing a correlation between the input content and a historical content to determine a search tag; (c) searching for related information in a database by utilizing the search tag and forming a feedback content; and (d) outputting the feedback content to the user. This invention may provide the user with a new search experience. The user may perform an interactive search and obtain a corresponding search result by means of natural speech or natural sentence expression.

Description

Dialogue search display method

The present invention relates to the technical field of data search, and in particular to a dialog-type search and display method that provides a user to freely search in a dialog to obtain a response.

Traditionally, if users want to search for data on the Internet, they usually perform search tasks through search engines.

The user enters the text to be searched in the search dialog box of the search engine, and the text is sent to the back-end data for comparison to return the related comparison results to the user; if the user wants to obtain accurate comparison results, then You need to enter the corresponding text and expressions according to the rules of the search engine. For users, you need to learn an additional input method. If you cross search in different search engines, the user may need to learn an additional input. Way, this is very inconvenient for the user.

In view of this, the present invention proposes a dialog-based search and display method to solve the lack of conventional technology.

The first object of the present invention is to provide a conversational search and display method, which analyzes the user's input content described in natural language or sentences, and responds to a feedback content according to the input content, so as to achieve the purpose of dialogue, search and display.

The second object of the present invention is to allow the user to provide input content through text, character strings, sounds, and gestures according to the aforementioned conversational search and display method, and further input content to further define the input content as a subject, a verb, a noun , A preposition, a subject, to achieve the purpose of conversion into search tags.

The third object of the present invention is based on the aforementioned conversational search display method. If the user cannot obtain appropriate feedback content through the search, the user will receive a search without results, a request for re-entry, and an unrecognized feedback content , And by establishing an initial information to achieve the purpose of self-learning.

The fourth object of the present invention is to record the user's input content according to the aforementioned dialog-based search and display method to understand the user's search behavior for subsequent marketing applications.

To achieve the above and other objectives, the present invention provides a dialogic search display method, which includes steps (a) retrieving an input content of a user; (b) parsing the input content and analyzing the input according to a natural language algorithm The relevance of the content and a historical content to determine a search tag, where the historical content is the input content that has been captured; (c) Use the search tag to search for a related information in a database and form a feedback content; And (d) output feedback content to the user.

Compared with the conventional technology, the present invention provides a dialog-based search and display method, so that users do not need to learn or adapt to the language and logic of the search engine, but use the most direct expression to search and obtain the required search results.

In order to fully understand the purpose, features and effects of the present invention, the following specific examples and the accompanying drawings are used to make a detailed description of the present invention, as follows:

In the present invention, "a" or "an" is used to describe the units, elements and components described herein. This is just for convenience of explanation, and provides a general meaning to the scope of the present invention. Therefore, unless clearly stated otherwise, such description should be understood to include one, at least one, and the singular also includes the plural.

In the present invention, the terms "including", "including", "having", "containing" or any other similar terms are intended to cover non-exclusive inclusions. For example, an element, structure, product, or device containing multiple elements is not limited to the elements listed here, but may include those not explicitly listed but which are generally inherent to the element, structure, product, or device. Other requirements. In addition, unless clearly stated to the contrary, the term "or" refers to an inclusive "or" rather than an exclusive "or".

Please refer to FIG. 1, which is a flowchart of a dialog search and display method according to a first embodiment of the present invention. In FIG. 1, the steps of the interactive search and display method are applied to an Internet or a local area network. In this embodiment, the Internet is taken as an example. The interactive search and display method can be applied between a personal electronic device and a server. For example, the electronic device can be a mobile communication device, a tablet computer, a personal computer, and the like. In a usage scenario, the server provides a search interface on the electronic device. A user logs in the server through the electronic device to obtain a search interface, and initiates a dialogic search and display with the server through the search interface.

The steps of the dialog-based search and display method start at step S11 and capture an input content of the user. In this step, the user can provide corresponding input content through the input unit of the electronic device, such as a keyboard, a mouse, a touch panel, a microphone, a camera, etc. to perform input actions; and, the input content can be, for example, a text or a string , A voice (voice, music, etc.) and a video (such as sign language, gestures, gestures, etc.). For example, the user can input text, character strings, patterns, etc. through the touch panel, the user can input a sound through the microphone, or the user can capture a video (such as sign language, etc.) through the camera. According to the aforementioned input method, the input action can be further divided into text type and non-text type, for example, text and string are text type, and sound and video are non-text type.

In another embodiment, after step S11, at least one of the electronic device and the server can convert the user's non-text or non-string input content to form input content having a text type or a string type .

Then, in step S12, an input content is analyzed according to a natural language algorithm and the correlation between the input content and a historical content is analyzed to determine a search tag. Among them, the historical content is the input content that has been captured. In this step, when the input content is converted into a search tag, the original unintentional text is changed into a text or string that can be understood by the system.

The foregoing analysis method can further define the input content of a subject, an adjective, a verb, a noun, a preposition, a subject, or natural language algorithms use a type of neural network to find the relevance of the input content Regardless of the method used, it is mainly used to parse the meaning of the input content.

In the process of determining the meaning of the input content, the historical content (one or more input content) is also judged together to confirm the meaning of the existing input content. Such a method can achieve a more interactive way of real dialogue between users. For example, the user inputs the input content of "Taiwan's highest mountain?" through a microphone, and the natural language algorithm further analyzes the input content, and parses "Taiwan's highest mountain?" into "Taiwan", "Highest", "Mountain", "Yes" and "?" "Taiwan" is defined as the main word, "highest" is defined as the adjective, "shan" is defined as the subject, "yes" is defined as the verb and "?" is defined as the interrogative sentence. According to the natural language algorithm, the traditional search engine can be regarded as a simple text input into an understandable sentence, that is, it can be understood in this step. The user needs to ask is that the geographic name is Taiwan, and in Taiwan Among the many mountains in the range, which is the highest altitude mountain? It is worth noting that the "Taiwan's highest mountain?" has become historical content after analysis.

Then, after receiving the result of "Taiwan's highest mountain is Yushan", the user further input "What about the world?" through the microphone. Although the second query ignored the word "mountain", the natural language In addition to parsing the input content, the algorithm also analyzes the correlation between the input content and historical content ("What is the highest mountain in Taiwan?"). Therefore, the natural language algorithm further speculates that the user is asking "Taiwan's highest mountain is Yushan, and which is the highest mountain in the world!", so by converting the input content into a search tag, in the conversation In the search and display method, it can search for qualified data from the data of many mountain ranges in the world, and then select "Himalayas". In this way, the dialog-based search and display method can complete the dialog-based search and display of the user's response to the query.

In addition to the aforementioned methods, the user asks "Which is the highest mountain in Taiwan?" It can also find the relevance of the input content through a neural network, such as searching the mountain database through the Internet, sharing texts, etc., to directly understand "Taiwan What is the highest mountain?"

Then in step S13, the search tag is used to search for a piece of related information in a database, and a feedback content is formed. Among them, the database pre-creates the information related to the search tag, continuing the previous example, creating the geography, type, altitude, altitude and other data in the database. By interpreting the search tag, the user needs to ask is that in Taiwan Among the many mountains in the range, which is the highest altitude mountain? Therefore, the database will find the qualified information Taiwan (geography), type (high mountain), and altitude (8,848m) according to the understood semantics, and then select "Yushan". The aforementioned "Yushan" information is packaged as feedback content.

In another embodiment, the search tag may search the database for the same or synonymous information, and form feedback content based on the information.

Following step S14, the feedback content is output to the user. In this embodiment, the feedback content is output to the electronic device.

Please refer to FIG. 2, which is a schematic flowchart of a dialog search and display method according to a second embodiment of the present invention. In FIG. 2, the steps of the interactive search and display method include steps S21 and S22 in addition to steps S11 to S14 of the first embodiment.

The steps S11 to S14 are as described in the first embodiment, and are not repeated here.

Step S21 is that after step S13, the search tag cannot search the database for the same or synonymous information. It responds to a search with no results, a request for re-entry, and an unrecognized feedback content. In this step, the user is informed by the feedback content that there is no corresponding search result.

Step S22 is to record the search tag after step S21 to create an initial message. For example, the creation method may be manual creation or automatic creation. For example, the automatic creation method can use the search tag to search for a network message on the Internet through a search engine, and then use the collected data to further perform the steps of filtering, selection, and analysis to create the initial information. .

The present invention has been disclosed in the above with preferred embodiments, but those skilled in the art should understand that this embodiment is only used to depict the present invention and should not be interpreted as limiting the scope of the present invention. It should be noted that all changes and substitutions equivalent to this embodiment should be included within the scope of the present invention. Therefore, the scope of protection of the present invention shall be subject to those defined in the scope of patent application.

S11-S14‧‧‧Method steps S21-S22‧‧‧Method steps

FIG. 1 is a schematic flowchart of a dialog search and display method according to a first embodiment of the invention. FIG. 2 is a schematic flowchart of a dialog-based search and display method according to a second embodiment of the invention.

S11-S14‧‧‧Method steps

Claims (9)

A conversational search and display method, including: (a) retrieving an input content of a user; (b) parsing an input content and analyzing the relevance of the input content and a historical content according to a natural language algorithm to Decide on a search tag, where the historical content is the input content that has been captured; (c) Use the search tag to search for a piece of related information in a database and form a feedback content; and (d) Output the feedback Content for this user. The conversational search and display method as described in item 1 of the patent application scope, wherein in step (a), the input content of at least one of a text, a string, a sound and a video of the user is captured, Wherein the text and the character string are in a text type, and the sound and the video are in a non-text type. The conversational search and display method as described in item 2 of the patent application scope, wherein after step (a), the input content of the non-text type of the user is converted to form the input content with a text type. Dialogue search and display method as described in item 1 or 3 of patent scope, wherein in step (b), the input content is parsed by a natural language algorithm to further define the input content as a subject, a verb , The search tag of at least one of an adjective, a noun, a preposition, and a subject. The dialog-based search and display method as described in item 4 of the patent application scope, wherein in step (c), the search tag may search the database for the same or synonymous information, and form the feedback content based on the information. The dialog-based search and display method as described in item 4 of the patent application scope, wherein after step (c), step (e) is included if the search tag cannot search the database for the same or synonymous information, reply A search no result, a request to re-enter the feedback content of at least one of the unrecognized. The dialog-based search and display method as described in item 6 of the patent application scope, wherein after step (e), step (f) is further included, and the search tag is recorded to create an initial message. The conversational search and display method as described in item 7 of the patent application scope, wherein in step (f), the search tag searches a network message on an Internet through a search engine to create the initial information. The conversational search and display method as described in item 1 of the patent application scope, wherein in step (b), the natural language algorithm uses a type of neural network to find relevance related to the input content.

Images