CN109753560B - Information processing method and device of intelligent question-answering system - Google Patents
Information processing method and device of intelligent question-answering system Download PDFInfo
- Publication number
- CN109753560B CN109753560B CN201910007508.7A CN201910007508A CN109753560B CN 109753560 B CN109753560 B CN 109753560B CN 201910007508 A CN201910007508 A CN 201910007508A CN 109753560 B CN109753560 B CN 109753560B
- Authority
- CN
- China
- Prior art keywords
- question
- user
- global variable
- standard
- user question
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3335—Syntactic pre-processing, e.g. stopword elimination, stemming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an information processing method and device of an intelligent question answering system. An information processing method of an intelligent question-answering system comprises the following steps: performing similarity calculation based on a first user question input by a user to determine a first standard question from the set of standard questions that matches the first user question; judging whether a word belonging to a global variable exists in the first user question; in response to the existence of a word belonging to a global variable, performing a similarity calculation based on the word belonging to the global variable in the first user question and a subsequent second user question to determine a second standard question matching the second user question; and outputting the answer associated with the second question as an answer to the second user question. According to the invention, the comprehension capability of the intelligent question-answering system to the question of the user is improved, and the user experience is improved.
Description
The invention provides a divisional application named as an information processing method and device of an intelligent question answering system, which has an application date of 2016, 11, month and 4 and an application number of 201610972523.1.
Technical Field
The invention relates to the technical field of man-machine interaction, in particular to an information processing method and device for an intelligent question-answering system.
Background
Human-computer interaction is the science of studying the interactive relationships between systems and users. The system may be a variety of machines, and may be a computerized system and software. For example, various artificial intelligence systems, such as intelligent customer service systems, voice control systems, and the like, may be implemented through human-computer interaction. Artificial intelligence semantic recognition is the basis for human-machine interaction, which is capable of recognizing human language for conversion into machine-understandable language.
The intelligent question-answering system is a typical application of human-computer interaction, wherein when a user proposes a question, the intelligent question-answering system gives an answer to the question. For this purpose, the intelligent question-answering system has a knowledge base in which a large number of questions and answers corresponding to each question are stored. The intelligent question-answering system firstly needs to identify the question provided by the user, namely, to find the question corresponding to the user question from the knowledge base, and then to find the answer matched with the question.
Current intelligent question-answering systems do not take into account the context of the context in which the user understands in isolation between questions presented in succession. For example, the user may have previously asked "the score line of enrollment in Jiangsu province this year ago," the latter sentence may ask "how many people we are enrolled outside this year," the intelligent question-and-answer system has no way of knowing at which province how many people are enrolled when understanding the latter sentence, and further needs to ask at which province. This affects the user experience.
Disclosure of Invention
The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects, and is intended to neither identify key or critical elements of all aspects nor delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that is presented later.
The invention provides an information processing method and system of an intelligent question-answering system, which aim to solve the problem that the intelligent question-answering system has low comprehension capability on question sentences of users.
In one aspect, the present invention provides an information processing method for an intelligent question-answering system, where the intelligent question-answering system includes a question-answering database, the question-answering database includes a set of standard questions, each standard question has a corresponding answer, and the information processing method includes:
performing similarity calculation based on a first user question input by a user to determine a first standard question from the set of standard questions that matches the first user question;
judging whether a word belonging to a global variable exists in the first user question;
in response to the existence of a word belonging to a global variable, performing a similarity calculation based on the word belonging to the global variable in the first user question and a subsequent second user question to determine a second standard question matching the second user question; and
outputting an answer associated with the second standard question as a response to the second user question;
the performing similarity calculation based on the words belonging to the global variable in the first user question and a subsequent second user question to determine a second standard question matching the second user question comprises:
performing semantic similarity calculation on the second user question and each semantic expression in the question-answer database to determine a first semantic expression with the highest similarity;
performing semantic similarity calculation on the combination of the words belonging to the global variables in the first user question and the second user question and each semantic expression in the question-answer database to determine a second semantic expression with the highest similarity;
if the similarity between the combination of the words belonging to the global variables in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, judging whether the part of speech corresponding to the words belonging to the global variables in the second semantic expression is the part of speech marked as the global variables, and if so, determining the standard question represented by the second semantic expression as the second standard question matched with the second user question.
Optionally, the performing the similarity calculation based on the first user question input by the user includes:
and performing semantic similarity calculation on the first user question and each semantic expression in the question-answer database, wherein the standard question represented by the semantic expression with the highest similarity is determined as the first standard question matched with the first user question.
Optionally, the determining whether a word belonging to a global variable exists in the first user question includes:
judging whether the semantic expression with the highest similarity to the first user question contains the part of speech marked as the global variable, if so, determining that the words corresponding to the part of speech of the global variable in the first user question are the words belonging to the global variable.
Optionally, whether the part of speech is marked as a global variable is preset according to actual requirements.
Optionally, parts of speech associated with a geographic location or time date are tagged as global variables.
Optionally, if the part of speech corresponding to the word belonging to the global variable in the second semantic expression is not the part of speech marked as the global variable, the standard question represented by the first semantic expression is determined as the second standard question matched with the second user question.
In another aspect, the present invention further provides an information processing apparatus of an intelligent question-answering system, where the intelligent question-answering system includes a question-answering database, the question-answering database includes a set of standard questions, each standard question has a corresponding answer, and the information processing apparatus includes:
a matching module for performing similarity calculation based on a first user question input by a user to determine a first standard question from the standard question set, the first standard question matching the first user question;
the global variable judging module is used for judging whether words belonging to global variables exist in the first user question sentence;
in response to the presence of a word belonging to a global variable, the matching module performs a similarity calculation based on the word belonging to the global variable in the first user question and a subsequent second user question to determine a second standard question that matches the second user question; and
an output module for outputting an answer associated with the second standard question as a response to the second user question;
the matching module includes:
a semantic similarity calculation module, configured to perform semantic similarity calculation on the second user question and each semantic expression in the question-and-answer database to determine a first semantic expression with the highest similarity, and perform semantic similarity calculation on a combination of a word belonging to a global variable in the first user question and the second user question and each semantic expression in the question-and-answer database to determine a second semantic expression with the highest similarity;
a judging module, configured to judge whether a similarity between a combination of a word belonging to a global variable in the first user question and the second semantic expression is greater than a similarity between the second user question and the first semantic expression, and further judge whether a part of speech corresponding to the word belonging to the global variable in the second semantic expression is a part of speech marked as the global variable when the similarity between the combination of the word belonging to the global variable in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, and if so, determine a standard question represented by the second semantic expression as the second standard question matched with the second user question.
Optionally, the semantic similarity calculation module is further configured to perform semantic similarity calculation on the first user question and each semantic expression in the question-answer database, where a standard question represented by a semantic expression with the highest similarity is determined as the first standard question matched with the first user question.
Optionally, the global variable determining module determines whether a semantic expression with the highest similarity to the first user question includes a part of speech marked as a global variable, and if so, a word in the first user question corresponding to the part of speech of the global variable is a word belonging to the global variable.
Optionally, when the part of speech corresponding to the word belonging to the global variable in the second semantic expression is not the part of speech marked as the global variable, the determining module determines the standard question represented by the first semantic expression as the second standard question matched with the second user question.
Compared with the prior art, the invention has the following beneficial effects:
by introducing the global variable, the intelligent question-answering system fully considers the context of the context to understand the question of the user, so that the comprehension capability of the question of the user is improved, and the user experience is improved.
Drawings
The above features and advantages of the present disclosure will be better understood upon reading the detailed description of embodiments of the disclosure in conjunction with the following drawings. In the drawings, components are not necessarily drawn to scale, and components having similar relative characteristics or features may have the same or similar reference numerals.
FIG. 1 is a flow chart illustrating an information processing method for an intelligent question answering system according to an aspect of the present invention; and
fig. 2 is a block diagram illustrating an information processing apparatus for an intelligent question answering system according to an aspect of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. It is noted that the aspects described below in connection with the figures and the specific embodiments are only exemplary and should not be construed as imposing any limitation on the scope of the present invention.
The basic knowledge points in the knowledge base are in the most primitive and simplest form the FAQs that are commonly used in ordinary times, and the common form is a question-answer pair. In the invention, the 'standard question' is a word for representing a certain knowledge point, and the main aim is to clearly express and facilitate maintenance. For example, "the tariff for a coloring ring back tone" is a clear description of the standard. The term "question" should not be construed narrowly as "question" but rather broadly as "input" with corresponding "output". For example, for semantic recognition for a control system, an instruction of a user, for example "turn on radio", should also be understood as a "question", in which case the corresponding "answer" may be a call to a control program for executing the corresponding control.
When the user inputs the information to the machine, the most ideal situation is to use standard questions, and the intelligent semantic recognition system of the machine can immediately understand the meaning of the user. However, rather than using standard questions, users often use some variant form of standard questions. For example, if the standard form of a station switch for a radio is "change station", then the command that the user may use is "switch station", and the machine also needs to be able to recognize that what the user has expressed is the same meaning.
Therefore, for intelligent semantic recognition, there is a need in the knowledge base for an expanded query of standard queries that is slightly different from the standard query expression but expresses the same meaning. The extended questions mainly appear in the form of semantic expressions. A standard question may be characterized by several semantic expressions.
The semantic expression mainly comprises words, parts of speech and their or relations, the core of the semantic expression depends on the parts of speech, the simple understanding of the parts of speech is a group of common words, the words can be similar or dissimilar in semantics, and the words can be marked as important or unimportant. The part of speech is a summary of a group of related words, and the part of speech is composed of a word name and a group of related words. The word class name is a word having a tag role in the group of related words, i.e., a representation of the word class. A part of speech contains at least one word (i.e., the part of speech itself).
The semantic expression and user question relationship is very different from the traditional template matching, in which the template and user question are just matching and unmatching relationships, and the relationship between the semantic expression and user question is represented by a quantized value (similarity), and the quantized value and the similarity between the similar question and the user question can be compared with each other. The following describes the specific composition of the semantic expression and the representation of the symbols.
Representation of a word ([ 2 ]])
To distinguish words from parts of speech in an expression, it is specified that the parts of speech must appear in brackets "[ ]", and the parts of speech appearing in brackets are generally "narrowly defined parts of speech", but can also be supported by configuring system parameters.
The following are some examples of simple expressions:
[ how the letter is opened ]
[ introduction ] [ multimedia message ] [ service ]
[ Feixin ] [ Login ] [ method ]
[ incoming call reminder ] [ how to ] [ charging ]
Or representation of a relationship (|)
The parts of speech in brackets may appear multiple times through "or" relationships whose parts of speech are computed separately in an "expanded" manner when computing the degree of similarity. "expansion" is a process of expanding a semantic expression into multiple simple expressions, primarily according to the meaning of "or". Such as: the method steps of the CRBT opening can be expanded into two simple semantic expressions of the steps of the CRBT opening and the method of the CRBT opening.
Examples of such semantic expressions are as follows:
[ method | step ] of [ CRBT ] opening ]
How to inquire and know PUK code
[ unsubscribe | Undo | close | Disable ] [ IP |17951] [ national Long distance discount packet ]
[ incoming call reminder ] [ function fee | monthly fee | information fee | communication fee ]
Unnecessary representations (
Parts of speech in brackets may be added at the end "? "indicates that there may or may not be a relationship, i.e., an unnecessary relationship, and the parts of speech of such an unnecessary relationship are also calculated separately in an" expanded "manner when calculating the similarity. "expansion" is mainly the process of expanding the unnecessary part of speech (or "or combination" of part of speech) contained in the semantic expression into two simple semantic expressions containing and not containing this part of speech. Such as: [ introduction ] [ mobile phone video ] [ military column ] [ what? The method can be expanded into two simple semantic expressions of 'introduction ] [ mobile phone video ] [ military column ] [ content ]' and 'introduction ] [ mobile phone video ] [ military column ] [ content ] [ what ]'.
Examples of such semantic expressions are as follows:
[ method | step of [ polyphonic ringtone ] [ cancellation ]? ]
[ introduction ] [ mobile phone video ] [ military column ] [ what? ]
[ introduction ] [ 12580? [ Life broadcasting ] [ quality and life edition ] [ free of charge ] [ business? ]
How to open ] [ mobile data | flow | internet ] [100 yuan ] [ package? [ short message ]
In the invention, the concept of global variables is introduced to the part of speech in the semantic expression. When performing semantic recognition, if a user question is positioned to a certain semantic expression through semantic similarity calculation, and the semantic expression has a part of speech set as a global variable, a word in the user question corresponding to the part of speech of the global variable is transmitted to a subsequent user question as a global variable in an interactive session between the intelligent question-answering system and the user to participate in the semantic similarity calculation.
FIG. 1 is a flow chart illustrating an information processing method 100 for an intelligent question and answer system in accordance with an aspect of the present invention. The intelligent question-answering system can comprise a question-answering database which contains a set of standard questions, each standard question having a corresponding answer. The information processing method 100 may include the steps of:
step 101: a similarity calculation is performed based on a first user question input by the user to determine a first standard question from the set of standard questions that matches the first user question.
In the standard question set, each standard question is associated with at least one semantic expression, and each standard question is characterized by the semantic expressions associated with the standard question. The semantic expression is used for carrying out similarity calculation with a user question input by a user so as to determine a standard question matched with the user question.
Specifically, semantic similarity calculation may be performed on the first user question and each semantic expression in the question-answer database, and the standard question represented by the semantic expression with the highest similarity is determined as the first standard question matching the first user question.
For example, the first user question asked by the user: the score line of the enrollment in Jiangsu province this year.
And executing semantic similarity calculation on the first user question and each semantic expression of each standard question in the question-answer database, and supposing that the semantic expression with the highest semantic similarity is found as follows:
[ coverage | declaration | admission ] [ school | Shanghai national language university | university? ] [ this year? Paper (test province)
Assuming that the standard questions corresponding to the semantic expression are: the upper and lower specific score lines;
assume that the answer to the standard question is: xx is the year's enrollment score line in xx. The answer is an application that may be passed parameters extracted from the user's question.
Step 102: and judging whether the words belonging to the global variable exist in the first user question sentence.
In the present invention, the determination is made by means of a semantic expression having the highest semantic similarity to the first user question. Specifically, whether the semantic expression with the highest similarity to the first user question contains the part of speech marked as the global variable is judged, and if yes, the word corresponding to the part of speech of the global variable in the first user question is the word belonging to the global variable.
Taking the semantic expression with the highest semantic similarity with the first user question as an example, assuming that the part of speech [ test province ] in the semantic expression is marked as a global variable, determining the words corresponding to the part of speech in the first user question as the words belonging to the global variable. For example, the word "jiangsu province" in the first user question corresponds to the part of speech [ examination province ], and is thus determined to be a word belonging to the global variable.
Whether the part of speech is marked as a global variable or not can be preset by an administrator of the system according to actual requirements. Parts of speech associated with, for example, geographic location, time-date, etc., are often tagged as global variables.
Step 103: in response to the existence of a word belonging to the global variable, a similarity calculation is performed based on the word belonging to the global variable in the first user question and a subsequent second user question to determine a second standard question matching the second user question.
Once a word in a user question is determined to be a word belonging to the global variable, that word is used for semantic similarity calculation of subsequent user questions. If there is no word belonging to the global variable, the similarity calculation of the subsequent user question only uses the subsequent user question itself to execute the semantic similarity calculation.
In one example, semantic similarity calculation is performed on the second user question and each semantic expression in the question-answer database to determine the first semantic expression with the highest similarity. In addition, semantic similarity calculation is performed on the combination of the words belonging to the global variables in the first user question and the second user question and each semantic expression in the question-answer database to determine a second semantic expression with the highest similarity.
And then judging whether the similarity between the combination of the words belonging to the global variables in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, if so, determining the standard question represented by the second semantic expression as the second standard question matched with the second user question, and if not, determining the standard question represented by the first semantic expression as the second standard question matched with the second user question.
As described above, the term determined as the global variable may contribute to semantic similarity calculation of a subsequent user question, so that the term and the subsequent user question are combined together as a whole to perform semantic similarity calculation once, if a semantic similarity result higher than that calculated by using the subsequent user question alone is obtained, it indicates that the term of the global variable improves accuracy of standard question positioning, and all standard questions positioned by using the combination of the term and the subsequent user question are used as matched standard questions to determine answers returned to the user.
On the contrary, if no semantic similarity result higher than that obtained by the semantic similarity calculation using the subsequent user question alone is obtained, the result indicates that the word of the global variable is not associated with the subsequent second user question, but the accuracy of the standard question positioning is reduced, so that the standard question positioned by using the subsequent user question alone is still used as the matched standard question to determine the answer returned to the user.
Continuing with the above example, assume that the second user question of the user's subsequent query is: in addition, we are not inviting people.
And respectively carrying out semantic similarity calculation twice on the second user question and the combination of the second user question and the words of the global variable of Jiangsu province and the second user question. It was found that the combination of a word using the global variable "Jiangsu province" and a second user question is targeted to the semantic expression:
[ university of Shanghai foreign language | school | university ] not recruited [ several | how many? [ the number of people? ] [ this year? Paper (test province)
The standard questions corresponding to the expression are: specific enrollment plan for Shanghai
The answer corresponding to the standard question is: xx present day. The answer is an application that may be passed parameters extracted from the user's question.
And the semantic expression has the highest semantic similarity, so the corresponding standard question is taken as the matched standard question, and the answer is output according to the standard question.
In the above example, it is determined whether the similarity between the combination of the words belonging to the global variable in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, and if so, the standard question represented by the second semantic expression is determined as the second standard question matched with the second user question. However, in another example, if the similarity between the combination of the word belonging to the global variable in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, further determination is still needed.
Specifically, if the similarity between the combination of the word belonging to the global variable in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, it is further determined whether the part of speech corresponding to the word belonging to the global variable in the second semantic expression is the part of speech marked as the global variable, if so, the standard question represented by the second semantic expression is determined as the second standard question matched with the second user question, otherwise, the standard question represented by the first semantic expression is determined as the second standard question matched with the second user question.
For example, assume that the second user question of the user's subsequent query is: how is the weather today.
Using only the second user question, then, the first semantic expression is located:
weather of [ date ]
The standard questions corresponding to the expression are: weather (weather)
The answer corresponding to the standard question is: what provinces of weather. The answer is a question-back sentence for asking the user which province is the weather, so as to give a targeted answer in the following.
And respectively carrying out semantic similarity calculation twice on the second user question and the combination of the second user question and the words of the global variable of Jiangsu province and the second user question. It was found that the combination of a word using the global variable "Jiangsu province" and a second user question is localized to a second semantic expression:
weather (date) (name of province of weather)
The standard questions corresponding to the expression are: provincial weather
The answer corresponding to the standard question is: xx province city weather today xx. The answer is an application that may be passed parameters extracted from the user's question.
In the above two semantic similarity calculations, the combination of the terms of the global variable of the second semantic expression and "Jiangsu province" and the second user question has a higher semantic similarity, and the first semantic expression and the second user question have a lower semantic similarity than the former.
At this time, it is necessary to further determine whether the part of speech [ weather province name ] corresponding to the word of the global variable "jiangsu province" is the global variable, if the part of speech [ weather province name ] is the global variable, the standard question represented by the second semantic expression is determined as the second standard question matched with the second user question, otherwise, if the part of speech [ weather province name ] is not the global variable, the standard question represented by the first semantic expression is determined as the second standard question matched with the second user question.
In this example, assuming that the part of speech [ weather province name ] is not a global variable, "weather" is still used as a matching standard question of the second user question sentence, and the corresponding answer is "weather of what province" at this time, that is, the system asks the user against which province's weather is being asked.
In this example, assuming that the part of speech [ weather province name ] is a global variable, "province weather" is used as a standard question for the second user to ask a sentence, and the corresponding answer is "xx" of the current weather in the city of xx province, that is, the system can directly give the weather in Jiangsu province through parameter calling.
Assuming that the above-mentioned intelligent question-answering system is mainly used for enrollment, for example, App for consulting policies of college entrance examination enrollment, global variables are set in semantic expressions of questions related to enrollment. When the user inquires the information of the enrollment, the user assumes that the inquired information is related to Jiangsu province, and the problems related to the enrollment can be automatically related to the Jiangsu province, so that the intelligent degree of the system is greatly improved. However, in such systems, global variables are not set for general queries, such as weather queries, because such questions asked by the user may not have continuity between questions as with solicited questions, in this way avoiding errors caused by automatically associating any question asked by the user with a global variable previously associated with the solicited question. For example, the user may ask weather other than Jiangsu, and possibly weather in Shanghai, and thus may cause erroneous feedback if the weather in Jiangsu is automatically returned.
While, for purposes of simplicity of explanation, the methodologies are shown and described as a series of acts, it is to be understood and appreciated that the methodologies are not limited by the order of acts, as some acts may, in accordance with one or more embodiments, occur in different orders and/or concurrently with other acts from that shown and described herein or not shown and described herein, as would be understood by one skilled in the art.
Fig. 2 is a block diagram illustrating an information processing apparatus 200 for an intelligent question answering system according to an aspect of the present invention. As shown in fig. 2, the intelligent question-answering system includes a question-answering database, which includes a set of standard questions, each of which has a corresponding answer. The information processing apparatus 200 may include a matching module 210, a global variable determination module 220, and an output module 230.
The matching module 210 may perform a similarity calculation based on a first user question input by the user to determine a first standard question from the set of standard questions that matches the first user question. The global variable determining module 220 may determine whether a word belonging to a global variable exists in the first user question.
Each criterion in the set of criteria has at least one semantic expression for characterizing the criterion, each semantic expression including at least one part of speech. In an example, the matching module 210 may include the semantic similarity calculation module 211 performing semantic similarity calculation on the first user question and each semantic expression in the question-answer database, where the standard question represented by the semantic expression with the highest similarity is determined as the first standard question matching the first user question. The global variable determining module 220 may determine whether the semantic expression with the highest similarity to the first user question includes a part of speech marked as a global variable, and if so, a word in the first user question corresponding to the part of speech of the global variable is a word belonging to the global variable.
Upon receiving a subsequent second user question, in response to a word belonging to the global variable existing in the first user question, the matching module 210 may perform a similarity calculation based on the word belonging to the global variable in the first user question and the subsequent second user question to determine a second standard question that matches the second user question. And if the words belonging to the global variable do not exist, performing similarity calculation only based on the second user question to determine a second standard question matched with the second user question.
Specifically, when there is a word belonging to the global variable in the first user question, the semantic similarity calculation module 211 may perform semantic similarity calculation on the second user question and each semantic expression in the question-and-answer database to determine the first semantic expression with the highest similarity, and perform semantic similarity calculation on a combination of the word belonging to the global variable in the first user question and the second user question and each semantic expression in the question-and-answer database to determine the second semantic expression with the highest similarity. The matching module 210 may further include a determining module 212, configured to determine whether a similarity between a combination of a word belonging to the global variable and the second user question in the first user question and the second semantic expression is greater than a similarity between the second user question and the first semantic expression, if so, determine the standard question represented by the second semantic expression as a second standard question matched with the second user question, and if not, determine the standard question represented by the first semantic expression as the second standard question matched with the second user question.
In another embodiment, when the similarity between the combination of the word belonging to the global variable in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, the determining module 212 further determines whether the part of speech corresponding to the word belonging to the global variable in the second semantic expression is the part of speech marked as the global variable, if so, determines the standard question represented by the second semantic expression as the second standard question matched with the second user question, and if not, determines the standard question represented by the first semantic expression as the second standard question matched with the second user question.
The output module 230 may output the answer associated with the second question as a response to the second user question.
The specific implementation manner of the information processing device of the intelligent question-answering system in the invention can be referred to the information processing method of the intelligent question-answering system, and is not described herein again.
According to the scheme of the invention, by introducing the global variable, the intelligent question-answering system fully considers the context of the context to understand the question of the user, so that the comprehension capability of the question of the user is improved, and the user experience is improved.
Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The previous description of the disclosure is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the disclosure is not intended to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. An information processing method of an intelligent question-answering system, wherein the intelligent question-answering system comprises a question-answering database, the question-answering database comprises a standard question set, each standard question has a corresponding answer, and the information processing method comprises the following steps:
performing similarity calculation based on a first user question input by a user to determine a first standard question from the set of standard questions that matches the first user question;
judging whether a word belonging to a global variable exists in the first user question;
in response to the existence of a word belonging to a global variable, performing a similarity calculation based on the word belonging to the global variable in the first user question and a subsequent second user question to determine a second standard question matching the second user question; and
outputting an answer associated with the second standard question as a response to the second user question;
the performing similarity calculation based on the words belonging to the global variable in the first user question and a subsequent second user question to determine a second standard question matching the second user question comprises:
performing semantic similarity calculation on the second user question and each semantic expression in the question-answer database to determine a first semantic expression with the highest similarity;
performing semantic similarity calculation on the combination of the words belonging to the global variables in the first user question and the second user question and each semantic expression in the question-answer database to determine a second semantic expression with the highest similarity;
if the similarity between the combination of the words belonging to the global variables in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, judging whether the part of speech corresponding to the words belonging to the global variables in the second semantic expression is the part of speech marked as the global variables, and if so, determining the standard question represented by the second semantic expression as the second standard question matched with the second user question.
2. The information processing method of claim 1, wherein the performing similarity calculation based on the first user question input by the user comprises:
and performing semantic similarity calculation on the first user question and each semantic expression in the question-answer database, wherein the standard question represented by the semantic expression with the highest similarity is determined as the first standard question matched with the first user question.
3. The information processing method according to claim 1, wherein the determining whether a word belonging to a global variable exists in the first user question includes:
judging whether the semantic expression with the highest similarity to the first user question contains the part of speech marked as the global variable, if so, determining that the words corresponding to the part of speech of the global variable in the first user question are the words belonging to the global variable.
4. The information processing method according to claim 1, wherein whether the part of speech is tagged as a global variable is preset according to an actual demand.
5. The information processing method according to claim 4, wherein a part of speech associated with a geographical location or a time date is marked as a global variable.
6. The information processing method according to claim 1, wherein if a part of speech corresponding to the word belonging to a global variable in the second semantic expression is not a part of speech tagged as a global variable, the standard question characterized by the first semantic expression is determined as the second standard question that matches the second user question.
7. An information processing apparatus of an intelligent question-answering system including a question-answering database including a set of standard questions each having a corresponding answer, the information processing apparatus comprising:
a matching module for performing similarity calculation based on a first user question input by a user to determine a first standard question from the standard question set, the first standard question matching the first user question;
the global variable judging module is used for judging whether words belonging to global variables exist in the first user question sentence;
in response to the presence of a word belonging to a global variable, the matching module performs a similarity calculation based on the word belonging to the global variable in the first user question and a subsequent second user question to determine a second standard question that matches the second user question; and
an output module for outputting an answer associated with the second standard question as a response to the second user question;
the matching module includes:
a semantic similarity calculation module, configured to perform semantic similarity calculation on the second user question and each semantic expression in the question-and-answer database to determine a first semantic expression with the highest similarity, and perform semantic similarity calculation on a combination of a word belonging to a global variable in the first user question and the second user question and each semantic expression in the question-and-answer database to determine a second semantic expression with the highest similarity;
a judging module, configured to judge whether a similarity between a combination of a word belonging to a global variable in the first user question and the second semantic expression is greater than a similarity between the second user question and the first semantic expression, and further judge whether a part of speech corresponding to the word belonging to the global variable in the second semantic expression is a part of speech marked as the global variable when the similarity between the combination of the word belonging to the global variable in the first user question and the second semantic expression is greater than the similarity between the second user question and the first semantic expression, and if so, determine a standard question represented by the second semantic expression as the second standard question matched with the second user question.
8. The information processing apparatus according to claim 7, wherein the semantic similarity calculation module is further configured to perform semantic similarity calculation on the first user question and each semantic expression in the question-and-answer database, and a standard question represented by a semantic expression with the highest similarity is determined as the first standard question that matches the first user question.
9. The information processing apparatus according to claim 7, wherein the global variable determination module determines whether a part of speech labeled as a global variable is included in the semantic expression having the highest similarity to the first user question, and if so, a word in the first user question corresponding to the part of speech of the global variable is a word belonging to the global variable.
10. The information processing apparatus according to claim 7, wherein the judgment module determines the standard question characterized by the first semantic expression as the second standard question that matches the second user question, when the part of speech corresponding to the word belonging to a global variable in the second semantic expression is not a part of speech tagged as a global variable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910007508.7A CN109753560B (en) | 2016-11-04 | 2016-11-04 | Information processing method and device of intelligent question-answering system |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610972523.1A CN106528759B (en) | 2016-11-04 | 2016-11-04 | The information processing method and device of intelligent Answer System |
| CN201910007508.7A CN109753560B (en) | 2016-11-04 | 2016-11-04 | Information processing method and device of intelligent question-answering system |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610972523.1A Division CN106528759B (en) | 2016-11-04 | 2016-11-04 | The information processing method and device of intelligent Answer System |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109753560A CN109753560A (en) | 2019-05-14 |
| CN109753560B true CN109753560B (en) | 2020-08-07 |
Family
ID=58349557
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910007508.7A Active CN109753560B (en) | 2016-11-04 | 2016-11-04 | Information processing method and device of intelligent question-answering system |
| CN201610972523.1A Active CN106528759B (en) | 2016-11-04 | 2016-11-04 | The information processing method and device of intelligent Answer System |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610972523.1A Active CN106528759B (en) | 2016-11-04 | 2016-11-04 | The information processing method and device of intelligent Answer System |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN109753560B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107239450B (en) * | 2017-06-02 | 2021-11-23 | 上海对岸信息科技有限公司 | Method for processing natural language based on interactive context |
| CN107256258B (en) | 2017-06-12 | 2019-09-06 | 上海智臻智能网络科技股份有限公司 | Semantic formula generation method and device |
| CN107807960B (en) * | 2017-09-30 | 2020-05-12 | 平安科技（深圳）有限公司 | Intelligent customer service method, electronic device and computer readable storage medium |
| CN107862005A (en) * | 2017-10-25 | 2018-03-30 | 阿里巴巴集团控股有限公司 | User view recognition methods and device |
| CN108536807B (en) * | 2018-04-04 | 2022-03-25 | 联想(北京)有限公司 | Information processing method and device |
| CN108536852B (en) * | 2018-04-16 | 2021-07-23 | 上海智臻智能网络科技股份有限公司 | Question-answer interaction method and device, computer equipment and computer readable storage medium |
| CN109189906A (en) * | 2018-08-17 | 2019-01-11 | 国家电网有限公司客户服务中心 | Intelligent customer service is to the complete semantic recognition methods of more question sentences under coherent context |
| CN109543020B (en) * | 2018-11-27 | 2022-11-04 | 科大讯飞股份有限公司 | Query processing method and system |
| CN110109937A (en) * | 2019-05-15 | 2019-08-09 | 北京市华风声像技术中心 | A kind of assessment of popular science knowledge and learning method, system and equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103198155A (en) * | 2013-04-27 | 2013-07-10 | 俞志晨 | Mobile terminal based smart question answering interaction system and method |
| CN103279528A (en) * | 2013-05-31 | 2013-09-04 | 俞志晨 | Question-answering system and question-answering method based on man-machine integration |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9158772B2 (en) * | 2012-12-17 | 2015-10-13 | International Business Machines Corporation | Partial and parallel pipeline processing in a deep question answering system |
| US20150169833A1 (en) * | 2013-12-16 | 2015-06-18 | Ulli Waltinger | Method and System for Supporting a Clinical Diagnosis |
| CN105677636A (en) * | 2015-12-30 | 2016-06-15 | 上海智臻智能网络科技股份有限公司 | Information processing method and device for intelligent question-answering system |
| CN105677822A (en) * | 2016-01-05 | 2016-06-15 | 首都师范大学 | Enrollment automatic question-answering method and system based on conversation robot |
-
2016
- 2016-11-04 CN CN201910007508.7A patent/CN109753560B/en active Active
- 2016-11-04 CN CN201610972523.1A patent/CN106528759B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103198155A (en) * | 2013-04-27 | 2013-07-10 | 俞志晨 | Mobile terminal based smart question answering interaction system and method |
| CN103279528A (en) * | 2013-05-31 | 2013-09-04 | 俞志晨 | Question-answering system and question-answering method based on man-machine integration |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106528759A (en) | 2017-03-22 |
| CN106528759B (en) | 2019-04-02 |
| CN109753560A (en) | 2019-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109753560B (en) | Information processing method and device of intelligent question-answering system | |
| CN110263144A (en) | A kind of answer acquisition methods and device | |
| CN106201424B (en) | A kind of information interacting method, device and electronic equipment | |
| CN107886948A (en) | Voice interactive method and device, terminal, server and readable storage medium storing program for executing | |
| CN109979457A (en) | A method of thousand people, thousand face applied to Intelligent dialogue robot | |
| CN106202301A (en) | A kind of intelligent response system based on degree of depth study | |
| CN108984650B (en) | Computer-readable recording medium and computer device | |
| WO2016201815A1 (en) | Method and apparatus for providing online customer service | |
| CN109065052B (en) | Voice robot | |
| CN105261366A (en) | Voice identification method, voice engine and terminal | |
| CN104462600A (en) | Method and device for achieving automatic classification of calling reasons | |
| CN112948534A (en) | Interaction method and system for intelligent man-machine conversation and electronic equipment | |
| CN111708869A (en) | Man-machine conversation processing method and device | |
| CN111400458A (en) | Automatic generalization method and device | |
| CN105630887A (en) | Representation method for knowledge markup languages of Chinese question answering system and Chinese question answering system | |
| CN111400465B (en) | Generation method and device of customer service robot, electronic equipment and medium | |
| CN110266900A (en) | Recognition methods, device and the customer service system that client is intended to | |
| CN109360565A (en) | A method of precision of identifying speech is improved by establishing resources bank | |
| CN109271503A (en) | Intelligent answer method, apparatus, equipment and storage medium | |
| TW202036323A (en) | Intelligent online customer service convergence core system which can provide a reply or query to an intention generated through voice recognition and natural language processing procedures | |
| CN109545203A (en) | Audio recognition method, device, equipment and storage medium | |
| CN105550361A (en) | Log processing method and apparatus, and ask-answer information processing method and apparatus | |
| CN112579031A (en) | Voice interaction method and system and electronic equipment | |
| CN110364155A (en) | Voice control error-reporting method, electric appliance and computer readable storage medium | |
| CN109901810A (en) | A kind of man-machine interaction method and device for intelligent terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230310 Address after: 200436 Floor 9, building a, yuncube, No. 1268, Wanrong Road, Jing'an District, Shanghai Patentee after: Huayuan computing technology (Shanghai) Co.,Ltd. Address before: 201803 7, 398 Lane 1555, Jiangxi Road, Jinsha, Jiading District, Shanghai. Patentee before: SHANGHAI XIAOI ROBOT TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |