CN110096692B - Semantic information processing method and device - Google Patents
Semantic information processing method and device Download PDFInfo
- Publication number
- CN110096692B CN110096692B CN201810081517.6A CN201810081517A CN110096692B CN 110096692 B CN110096692 B CN 110096692B CN 201810081517 A CN201810081517 A CN 201810081517A CN 110096692 B CN110096692 B CN 110096692B
- Authority
- CN
- China
- Prior art keywords
- semantic information
- conclusions
- extracting
- known conditions
- implicit
- 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- 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)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Machine Translation (AREA)
Abstract
The embodiment of the invention provides a semantic information processing method and a semantic information processing device, and belongs to the technical field of information processing. The semantic information processing method comprises the steps of dividing a question stem into two parts, namely a known condition and a conclusion, according to the obtained question stem; extracting the explicit semantic information in the known conditions and the conclusions according to the obtained known conditions and the conclusions; when implicit semantic information exists in the known conditions and/or conclusions, extracting the implicit semantic information in the known conditions and/or conclusions; and combining the extracted explicit semantic information and the extracted implicit semantic information to obtain the semantic information of the question stem. The semantic information processing method provided by the embodiment of the invention can extract more comprehensive and complete semantic information from the question stem, thereby providing accurate and uniform input information for automatic answer of the question.
Description
Technical Field
The embodiment of the invention relates to the technical field of information processing, in particular to a semantic information processing method and a semantic information processing device.
Background
In order to improve the efficiency of information processing and to comprehensively grasp necessary information, feature extraction is becoming a research hotspot.
A semantic-based feature extraction method is one of the commonly used feature extraction methods, and how to extract useful information from sentences with multiple expression modes becomes an urgent problem to be solved.
Disclosure of Invention
In view of this, one of the technical problems solved by the embodiments of the present invention is to provide a semantic information processing method and apparatus, which can accurately extract semantic information from a question stem of a question, so as to provide accurate and uniform input for automatic solution of the question.
In a first aspect, an embodiment of the present invention provides a semantic information processing method, including:
dividing the question stem into two parts of a known condition and a conclusion according to the obtained question stem;
according to the obtained known conditions and conclusions, extracting explicit semantic information in the known conditions and the conclusions;
when implicit semantic information exists in the known condition and/or conclusion, extracting the implicit semantic information in the known condition and/or conclusion;
and combining the extracted explicit semantic information and the extracted implicit semantic information to obtain the semantic information of the question stem.
Optionally, in a specific embodiment based on the first aspect, the step of determining that there is implicit semantic information in the known condition and/or conclusion is:
and according to the known condition and/or conclusion, judging that the known condition and/or conclusion has implicit semantic information when an entity is missing in the semantic information obtained according to the extraction mode of the explicit semantic information. Optionally, the step of extracting the explicit semantic information in the known conditions and conclusions according to the obtained known conditions and conclusions specifically includes:
obtaining the corresponding relation of the keywords in the known conditions and the conclusions according to the corresponding table of the keywords and the relation;
and extracting the explicit semantic information in the known conditions and the conclusions by combining the corresponding relation of the keywords according to the position relation between the keywords and the entities in the known conditions and the conclusions.
Optionally, when implicit semantic information exists in the known condition and/or conclusion, the specific step of extracting the implicit semantic information in the known condition and/or conclusion is as follows:
when an entity is missing in the semantic information obtained according to the extraction mode of the dominant semantic information, obtaining a keyword corresponding to the relation according to the relation in the semantic information with the missing entity;
and extracting implicit semantic information in the known conditions and/or conclusions according to statements before and/or after the keywords.
Optionally, the step of extracting the explicit semantic information in the known conditions and conclusions according to the obtained known conditions and conclusions specifically includes:
obtaining a single sentence in the known conditions and the conclusion which is divided by the comma according to the comma in the known conditions and the conclusion;
respectively extracting explicit semantic information from the single sentences;
and combining the explicit semantic information extracted from each single statement to obtain the explicit semantic information in the known conditions and the conclusions.
Optionally, the merging the extracted explicit semantic information and implicit semantic information to obtain the semantic information of the stem further includes:
and in the process of extracting the latent semantic information in the known conditions and/or conclusions according to the statements before and/or after the statement of the keyword, if the supplementary semantic information of the question stem is found, extracting the supplementary semantic information, and recording the supplementary semantic information as the latent semantic information.
In a second aspect, an embodiment of the present invention further provides a semantic information processing apparatus, including:
the dividing module is used for dividing the question stem into two parts of a known condition and a conclusion according to the obtained question stem;
the explicit semantic information extraction module is used for extracting the explicit semantic information in the known conditions and the conclusions according to the obtained known conditions and the conclusions;
the implicit semantic information extracting module is used for extracting the implicit semantic information in the known condition and/or conclusion when the known condition and/or conclusion has the implicit semantic information;
and the merging module is used for merging the explicit semantic information extracted by the explicit semantic information extraction module and the implicit semantic information extracted by the implicit semantic information extraction module to obtain the semantic information of the question stem. Optionally, in a specific embodiment of the present invention based on the second aspect, the apparatus further includes a determining module, configured to:
according to the known conditions and/or the conclusion, when entities are missing in the semantic information obtained according to the extraction mode of the explicit semantic information, judging that the known conditions and/or the conclusion have implicit semantic information;
the implicit semantic information extracting module is connected with the judging module and is used for extracting the implicit semantic information in the known conditions and/or the conclusions when the judging module judges that the implicit semantic information exists in the known conditions and/or the conclusions;
the merging module is specifically configured to merge the explicit semantic information extracted by the explicit semantic information extracting module and the implicit semantic information extracted by the implicit semantic information extracting module to obtain semantic information of the question stem.
Optionally, the explicit semantic information extracting module is specifically configured to: obtaining the corresponding relation of the keywords in the known conditions and the conclusions according to the corresponding table of the keywords and the relation;
and extracting the explicit semantic information in the known conditions and the conclusions by combining the corresponding relation of the keywords according to the position relation between the keywords and the entities in the known conditions and the conclusions.
Optionally, the implicit semantic information extracting module is specifically configured to:
when an entity is missing in the semantic information obtained according to the extraction mode of the dominant semantic information, obtaining a keyword corresponding to the relationship according to the relationship in the semantic information with the entity missing;
extracting implicit semantic information in the known condition and/or conclusion according to statements before and/or after the keyword, and,
and in the process of extracting the latent semantic information in the known conditions and/or conclusions according to the statements before and/or after the statement of the keyword, if the supplementary semantic information of the question stem is found, extracting the supplementary semantic information, and recording the supplementary semantic information as the latent semantic information.
According to the technical scheme, the embodiment of the invention can extract the comprehensive question stem information from the question stem, and the extracted question stem semantic information group can contain all information in the question stem, so that accurate, effective and uniform input is provided for automatic answer of the question.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present invention, and it is also possible for a person skilled in the art to obtain other drawings based on the drawings.
Fig. 1 is a schematic flow chart of a semantic information processing method according to an embodiment of the present invention;
fig. 2 is a device structure diagram of a semantic information processing device according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments of the present invention shall fall within the scope of the protection of the embodiments of the present invention.
Referring to fig. 1, an embodiment of the present invention provides a semantic processing method, including steps S100 to S400, specifically:
s100: and dividing the question stem into two parts of a known condition and a conclusion according to the obtained question stem.
It should be noted that the semantic information processing method provided in the embodiment of the present application is applicable to various question types including geometric certification questions, calculation questions, and the like, and the embodiment of the present application is described herein by taking extraction of semantic information in a geometric certification question stem as an example.
According to the embodiment of the application, through statistical analysis of a large number of proving question stem texts, the proving question stem can be divided into two parts of a known condition and a conclusion by means of 'proving' or 'proving', the part of the proving question stem before 'proving' or 'proving' is the known condition, and the part of the proving question stem after 'proving' or 'proving' is the conclusion. Therefore, in the embodiment of the present application, the known conditions and conclusions for proving the question stem can be obtained by traversing the question stem and finding "proving" or "proving".
Similarly, when extracting the question stem semantic information of the calculation question, the known conditions and conclusions of the question stem can also be classified through calculation, solution and the like.
It should be noted that the "semantic information" in the embodiment of the present application includes explicit semantic information and implicit semantic information. The explicit semantic information refers to semantic information which does not need to contact context and can be directly obtained according to a single statement and comprises two entities and a relationship between the two entities, wherein the two entities can be respectively distributed on two sides of the corresponding relationship, and the distance between each entity and the corresponding relationship is the shortest on each side, such as AB × CD; or distributed on the same side of the corresponding relationship, for example, AC is the diagonal line of diamond ABCD; implicit semantic information refers to information that can be obtained only by contacting the context, and for the geometric proof question, "semantic information" refers to mathematical information contained in the question stem sentence.
S200: and extracting the explicit semantic information in the known conditions and the conclusions according to the obtained known conditions and the conclusions.
In actual operation, optionally, obtaining a relation corresponding to the keyword in the known condition and the conclusion according to a corresponding table of the keyword and the relation; and extracting the explicit semantic information in the known conditions and the conclusions by combining the corresponding relation of the keywords according to the position relation between the keywords and the entities in the known conditions and the conclusions.
It should be noted that the entities in the embodiments of the present application refer to specific units of basic concepts in mathematics, such as a triangle ABC, a line segment AB, a point O, and the like; relationships refer to entities and relationships between entities, such as vertical, collinear, and the like. The semantic information refers to a set consisting of entities and relations, and can be represented by { entity 1, entity 2, relation between entity 1 and entity 2 }, for example, a statement with an AB × CD in a known condition or conclusion of a stem can be obtained, and the semantic information corresponding to the statement is { AB, CD, vertical }.
Optionally, according to a comma in the known condition and conclusion, obtaining a single sentence in the known condition and conclusion divided by the comma; respectively extracting explicit semantic information from the single sentences; and combining the explicit semantic information extracted from each single statement to obtain the explicit semantic information in the known conditions and the conclusions.
In actual implementation, a keyword capable of representing a relationship between entities can be found from single sentences in known conditions and/or conclusions segmented by commas, and the basis for finding the keyword can refer to table 1:
TABLE 1 keyword and relationship correspondence table
And after the relation corresponding to the key words in the proving question stem is obtained, extracting the explicit semantic information according to the relation and the position relation of the entity in the sentence. For example, if it is proved that a single statement in the known condition of the topic stem is "DE ≠ AB", then the keyword "″) can be obtained by traversing the statement" DE ″) AB ", the relationship corresponding to the keyword" ″) is obtained as vertical by looking up the table 1, and two sides of the keyword "″) are respectively provided with an entity, which is" DE "and" AB ", so that the dominant semantic information { DE, AB, vertical } or { AB, DE, vertical } corresponding to the statement" DE ″) AB "can be extracted according to the position relationship of the keyword" ″) and the entity in the statement "DE ″ AB".
S300: when implicit semantic information exists in the known condition and/or conclusion, extracting the implicit semantic information in the known condition and/or conclusion.
In the process of actually extracting semantic information of the question stem, after the relation key words are found, the explicit semantic information obtained by searching the entities on the two sides of the relation key words or the entity on the same side of the key words may not be the complete semantic information of the question stem, that is, the situation of implicit semantic information may also exist in the known conditions and/or the conclusion of the question stem.
Optionally, according to the known condition and/or conclusion, when an entity is missing in the semantic information obtained according to the extraction manner of the explicit semantic information, it is determined that the known condition and/or conclusion has implicit semantic information.
Further, when an entity is missing in the semantic information obtained according to the extraction mode of the dominant semantic information, obtaining a keyword corresponding to the relation according to the relation in the semantic information with the missing entity; and extracting implicit semantic information in the known conditions and/or conclusions according to statements before and/or after the keywords.
For example, if we obtain "as shown, the diagonals AC, BD of the diamond ABCD intersect at point O, BE/AC, CE/DB. for evidence: BE ×) ce ", in the embodiment of the present application, the stem is first divided into two parts of known conditions and conclusions according to" finding evidence ":
the known conditions are: as shown, the diagonals AC, BD of the diamond ABCD intersect at a point O, BE/AC, CE/DB.
And (4) conclusion: BE ≠ CE.
According to the commas in the above known conditions and conclusions, the single sentences in the known conditions separated by the commas are respectively obtained:
single statement a: as shown in the figure.
Single statement b: the diagonals AC, BD of the diamond ABCD intersect at point O.
A single statement c: BE/AC.
The single statement d: CE/DB.
And a single statement in the conclusion:
single statement e: BE ≠ CE.
Since the principle of extracting semantic information from known conditions is the same as that of extracting semantic information from conclusions, the embodiments of the present application only illustrate the extraction of semantic information from known conditions.
For a single statement a, through traversing the statement, the statement is found not to contain keywords capable of embodying the relationship, and therefore semantic information cannot be obtained from the single statement a.
For a single sentence b, by traversing the sentence, a diagonal is obtained, and the diagonal intersects two keywords capable of representing the relationship, and the collinear relationship can be obtained by intersecting the keyword, which is definitely corresponding to three entities, two semantic sets consisting of entities and relationships, while the diagonal does not necessarily correspond to three entities, two semantic sets consisting of entities and relationships, which are obtained from the keyword, because only one diagonal may be used in the question stem. According to the keyword and the corresponding table of the relationship, namely table 1 in the embodiment of the application, the obtained relationship corresponding to the diagonal line is the diagonal line, and the relationship intersecting with the diagonal line is collinear.
Furthermore, because the rhombus ABCD and the diagonal line AC are positioned at two sides of the diagonal line of the keyword, the dominant semantic information { rhombus ABCD, AC and diagonal line } is obtained according to the entity rhombus ABCD and the entity AC which are positioned at two sides of the diagonal line and have the shortest distance with the diagonal line; and semantic information obtained by extracting the explicit semantic information according to the information of the entity set consisting of three entities, two entities and a relation, which is obtained from the entity BD and the point O which intersect at two sides and have the shortest distance with the entity BD and the point O, and the information of the semantic set which is obtained from the entity BD and the relation and corresponds to the three entities, is { BD, point O, collinear }, and {? Point O, collinear }, where? Indicating the missing entity. At this time, it can be presumed that implicit semantic information exists in the single statement b, and the missing entity needs to be found by contacting the context and be supplemented to the corresponding semantic information.
From semantic information {? The relation in the semantic information obtained from the points O and collinear is collinear, and the keywords in the stem corresponding to the collinear are intersected and are taken as the starting point for searching the missing entity, the previous sentence is the diagonal line AC and BD of the rhombus ABCD, the next sentence is the point O, obviously, the next sentence does not contain the missing entity, therefore, the intersection is taken as the starting point, the previous sentence is searched according to the direction from right to left, the BD entity is searched first, at this time, whether the BD entity is the missing entity can be judged, and the BD as the missing entity can be directly supplemented into the semantic information with the missing entity.
On one hand, in the actual execution process, if the BD entity is to be determined as the missing entity, the BD entity may be determined by determining whether the BD is located at one side of the BD and is closest to the intersecting side, if so, the BD is not the missing entity, and if not, the BD may be supplemented as the missing entity to the semantic information of the missing entity.
On the other hand, in the actual execution process, if the BD is not determined to be the missing entity, the BD can be directly supplemented to the semantic information of the missing entity as the missing entity.
For the two aspects, after the BD is used as a missing entity and is supplemented into the semantic information of the missing entity, whether the supplemented complete semantic information is repeated with the extracted semantic information is checked, and if the supplemented complete semantic information is repeated, the conclusion that the BD is not the missing entity is obtained; if not, the BD is taken as the missing entity, and it is obvious that the BD is not the missing entity in this embodiment.
Based on the above, continue to search forward to find "before, except for the entity BD, the entity AC that is closest to the intersection is supplemented with AC {? Point O, collinearity, to get { AC, point O, collinearity }.
Optionally, in the process of extracting the latent semantic information in the known condition and/or conclusion according to the statements before and/or after the statement in which the keyword is located, if the supplementary semantic information of the stem is found, extracting the supplementary semantic information, and recording the supplementary semantic information as the latent semantic information;
and merging all the recessive semantic information and the extracted dominant semantic information, namely merging the dominant semantic information, and obtaining the semantic information of the question stem by firstly obtaining the recessive semantic information and then obtaining the supplementary semantic information.
As can be seen from the foregoing embodiments, in the process of searching for a missing entity, "is found, entities located at both sides of the pause number and closest to the pause number are AC and BD, respectively, a parallel phrase is divided according to the meaning represented by the pause number, that is, the pause number is used to obtain a relationship corresponding to AC and a relationship corresponding to BD, and a supplementary semantic information { diamond ABCD, AC, diagonal } can also be obtained according to the semantic information { diamond ABCD, BD, diagonal }.
Of course, the embodiment of the present application may also obtain implicit semantic information of { AC, point O, collineation } through the pause sign ", the entities AC, BD distributed at the pause sign", "at both sides, and the extracted semantic information { BD, point O, collineation }.
S400: and merging the extracted semantic information to obtain the semantic information of the question stem.
In summary, according to a single statement b, the explicit semantic information corresponding to the single statement b is { diamond ABCD, AC, diagonal }, { BD, point O, collinear } and the implicit semantic information is { AC, point O, collinear }, { diamond ABCD, BD, diagonal }, and the explicit semantic information and the implicit semantic information are combined to obtain the semantic information corresponding to the single statement b, which is { diamond ABCD, AC, diagonal }, { BD, point O, collinear }, { AC, point O, collinear } and { diamond ABCD, BD, diagonal }.
Similarly, the semantic information corresponding to the single statement c is { BE, AC, parallel }, the semantic information corresponding to the single statement d is { CE, DB, parallel }, and the semantic information corresponding to the single statement e is { BE, CE, vertical }.
As can BE seen from the above, the semantic information corresponding to the known condition in the stem is { diamond ABCD, AC, diagonal }, { diamond ABCD, BD, diagonal }, { BD, point O, collinear }, and { AC, point O, collinear }, { BE, AC, parallel }, { CE, DB, parallel }, { BE, CE, vertical }.
According to the same method for extracting the semantic information of the known conditions in the stem, the semantic information corresponding to the conclusion BE ≠ CE is { BE, CE, vertical }. Therefore, the semantic information corresponding to the known conditions and the semantic information corresponding to the conclusion are merged, and the semantic information of the stem can be obtained as follows:
the known conditions are: { rhombus ABCD, AC, diagonal }, { rhombus ABCD, BD, diagonal }, { BD, point O, collinear } and { AC, point O, collinear }, { BE, AC, parallel }, { CE, DB, parallel };
and (4) conclusion: { BE, CE, vertical }.
When the method is applied specifically, the semantic information of the question stem can be input into the automatic answering system, so that the automatic answering system can answer the question automatically according to the input information.
According to the semantic information processing method provided by the embodiment of the application, when the hidden semantic information exists in the question stem, the hidden semantic information can be extracted, and the entity and relation full coverage can be realized, so that comprehensive and complete semantic information can be extracted, and accurate and uniform input information is provided for automatic answer of the question.
Based on the same inventive concept, referring to fig. 2, an embodiment of the present application further provides a semantic information processing apparatus, including:
the dividing module 201 is configured to divide the question stem into two parts, namely a known condition and a conclusion, according to the obtained question stem;
the explicit semantic information extracting module 202 is configured to extract explicit semantic information in the known conditions and the conclusions according to the obtained known conditions and the conclusions;
the implicit semantic information extracting module 203 is configured to, when implicit semantic information exists in the known condition and/or conclusion, extract the implicit semantic information in the known condition and/or conclusion;
a merging module 204, configured to merge the explicit semantic information extracted by the explicit semantic information extracting module and the implicit semantic information extracted by the implicit semantic information extracting module to obtain semantic information of the stem.
Optionally, the apparatus further includes a determining module, configured to:
according to the known conditions and/or the conclusion, when entities are missing in the semantic information obtained according to the extraction mode of the explicit semantic information, judging that the known conditions and/or the conclusion have implicit semantic information;
the implicit semantic information extracting module 203 is connected to the judging module, and is configured to extract the implicit semantic information in the known condition and/or the conclusion when the judging module judges that the implicit semantic information exists in the known condition and/or the conclusion;
the merging module 204 is specifically configured to merge the explicit semantic information extracted by the explicit semantic information extracting module 202 and the implicit semantic information extracted by the implicit semantic information extracting module 203 to obtain the semantic information of the stem.
Optionally, the explicit semantic information extracting module 202 is specifically configured to:
obtaining the corresponding relation of the keywords in the known conditions and the conclusions according to the corresponding table of the keywords and the relation;
and extracting the explicit semantic information in the known conditions and the conclusions by combining the corresponding relation of the keywords according to the position relation between the keywords and the entities in the known conditions and the conclusions.
Optionally, the implicit semantic information extracting module 203 is specifically configured to:
when an entity is missing in the semantic information obtained according to the extraction mode of the dominant semantic information, obtaining a keyword corresponding to the relation according to the relation in the semantic information with the missing entity;
extracting implicit semantic information in the known condition and/or conclusion from statements before and/or after the keyword, and,
and in the process of extracting the latent semantic information in the known conditions and/or conclusions according to the statements before and/or after the statement of the keyword, if the supplementary semantic information of the question stem is found, extracting the supplementary semantic information, and recording the supplementary semantic information as the latent semantic information.
In the embodiment of the present application, the dividing module 201, the explicit semantic information extracting module 202, the implicit semantic information extracting module 203, and the merging module 204 may be configured to implement corresponding steps in the foregoing method embodiment.
It should be noted that, in the embodiment of the present application, the entity in the known condition and/or conclusion may also be obtained first, then the relation word in the known condition and/or conclusion is obtained, and then the explicit semantic information in the known condition and/or conclusion is extracted according to the obtained relative position between the relation word and the entity in the known condition and/or conclusion. That is, the embodiment of the present application does not specifically limit the order of obtaining the relation words and obtaining the entities from the known conditions and/or conclusions.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative functional modules and method steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. 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.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the division of the above-described modules or units is only one type of division of logical functions, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
The integrated unit may be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as a separate product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium and can implement the steps of the embodiments of the method when the computer program is executed by a processor. The computer program includes computer program code, and the computer program code may be in a source code form, an object code form, an executable file or some intermediate form. The computer-readable storage medium may include: any entity or device capable of carrying the above-mentioned computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signal, telecommunication signal, software distribution medium, etc. It should be noted that the computer readable storage medium may contain content that is subject to appropriate increase or decrease according to the requirements of legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable storage media does not include electrical carrier signals and telecommunication signals according to legislation and patent practice.
The above embodiments are only for illustrating the technical solutions of the present invention, and not for limiting the same. While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention. It will be apparent to those skilled in the art that various modifications and variations can be made in the embodiments of the present invention without departing from the spirit or scope of the embodiments of the invention. Thus, if such modifications and variations of the embodiments of the present invention fall within the scope of the claims of the embodiments of the present invention and their equivalents, the embodiments of the present invention are also intended to encompass such modifications and variations.
Claims (9)
1. A semantic information processing method is characterized by comprising the following steps:
dividing the question stem into two parts of a known condition and a conclusion according to the obtained question stem;
according to the obtained known conditions and conclusions, extracting explicit semantic information in the known conditions and the conclusions, wherein the explicit semantic information is semantic information which does not need to be connected with a context and can be directly obtained according to a single statement and comprises two entities and a relation between the two entities;
when implicit semantic information exists in the known conditions and/or the conclusions, extracting the implicit semantic information in the known conditions and/or the conclusions, wherein the implicit semantic information is semantic information obtained by needing to contact the context;
merging the extracted explicit semantic information and the extracted implicit semantic information to obtain semantic information of the question stem;
the judging step that the known conditions and/or conclusions have implicit semantic information comprises the following steps:
after extracting the explicit semantic information in the known conditions and the conclusions according to the obtained known conditions and the conclusions, judging that the known conditions and/or the conclusions have implicit semantic information when the obtained semantic information has entity missing.
2. The semantic information processing method according to claim 1, wherein the step of extracting the explicit semantic information in the known condition and conclusion according to the obtained known condition and conclusion specifically comprises:
obtaining the corresponding relation of the keywords in the known conditions and the conclusions according to the corresponding table of the keywords and the relation;
and extracting the explicit semantic information in the known conditions and the conclusions by combining the corresponding relation of the keywords according to the position relation between the keywords and the entities in the known conditions and the conclusions.
3. The semantic information processing method according to claim 1 or 2, wherein when implicit semantic information exists in the known condition and/or conclusion, the specific steps of extracting the implicit semantic information in the known condition and/or conclusion are as follows:
when an entity is missing in the semantic information obtained according to the extraction mode of the dominant semantic information, obtaining a keyword corresponding to the relation according to the relation in the semantic information with the missing entity;
and extracting implicit semantic information in the known conditions and/or conclusions according to statements before and/or after the keywords.
4. The semantic information processing method according to claim 1, wherein the step of extracting the explicit semantic information in the known condition and conclusion according to the obtained known condition and conclusion specifically comprises:
obtaining a single sentence in the known conditions and the conclusion which is divided by the comma according to the comma in the known conditions and the conclusion;
respectively extracting explicit semantic information from the single sentences;
and combining the explicit semantic information extracted from each single statement to obtain the explicit semantic information in the known conditions and the conclusions.
5. The semantic information processing method according to claim 3, wherein the merging the extracted explicit semantic information and implicit semantic information to obtain the semantic information of the stem further comprises:
and in the process of extracting the latent semantic information in the known conditions and/or conclusions according to the statements before and/or after the statement of the keyword, if the supplementary semantic information of the question stem is found, extracting the supplementary semantic information, and recording the supplementary semantic information as the latent semantic information.
6. A semantic information processing apparatus characterized by comprising:
the dividing module is used for dividing the question stem into two parts of a known condition and a conclusion according to the obtained question stem;
the explicit semantic information extraction module is used for extracting the explicit semantic information in the known conditions and the conclusions according to the obtained known conditions and the conclusions, wherein the explicit semantic information is semantic information which does not need to be connected with a context and can be directly obtained according to a single statement and comprises two entities and a relation between the two entities;
the implicit semantic information extracting module is used for extracting the implicit semantic information in the known condition and/or conclusion when the known condition and/or conclusion has the implicit semantic information, wherein the implicit semantic information is the semantic information obtained by needing to contact the context;
the merging module is used for merging the explicit semantic information extracted by the explicit semantic information extracting module and the implicit semantic information extracted by the implicit semantic information extracting module to obtain the semantic information of the question stem;
the device further comprises a judging module for:
after extracting the explicit semantic information in the known conditions and the conclusions according to the obtained known conditions and the conclusions, judging that the known conditions and/or the conclusions have implicit semantic information when the obtained semantic information has entity missing.
7. The semantic information processing apparatus according to claim 6, characterized in that;
the implicit semantic information extracting module is connected with the judging module and used for extracting the implicit semantic information in the known conditions and/or the conclusions when the judging module judges that the implicit semantic information exists in the known conditions and/or the conclusions.
8. The semantic information processing apparatus according to claim 6 or 7, wherein the explicit semantic information extracting module is specifically configured to:
obtaining the corresponding relation of the keywords in the known conditions and the conclusions according to the corresponding table of the keywords and the relation;
and extracting the explicit semantic information in the known conditions and the conclusions by combining the corresponding relation of the keywords according to the position relation between the keywords and the entities in the known conditions and the conclusions.
9. The semantic information processing apparatus according to claim 6 or 7, wherein the implicit semantic information extraction module is specifically configured to:
when an entity is missing in the semantic information obtained according to the extraction mode of the dominant semantic information, obtaining a keyword corresponding to the relation according to the relation in the semantic information with the missing entity;
extracting implicit semantic information in the known condition and/or conclusion from statements before and/or after the keyword, and,
and in the process of extracting the latent semantic information in the known conditions and/or conclusions according to the statements before and/or after the statement of the keyword, if the supplementary semantic information of the question stem is found, extracting the supplementary semantic information, and recording the supplementary semantic information as the latent semantic information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810081517.6A CN110096692B (en) | 2018-01-29 | 2018-01-29 | Semantic information processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810081517.6A CN110096692B (en) | 2018-01-29 | 2018-01-29 | Semantic information processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110096692A CN110096692A (en) | 2019-08-06 |
| CN110096692B true CN110096692B (en) | 2021-04-09 |
Family
ID=67442648
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810081517.6A Active CN110096692B (en) | 2018-01-29 | 2018-01-29 | Semantic information processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110096692B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340662A (en) * | 2020-02-25 | 2020-06-26 | 上海乂学教育科技有限公司 | Geometric certification question making system |
| CN117494674B (en) * | 2023-11-15 | 2024-07-16 | 华中师范大学 | Automatic geometric proof question generation method and system based on existing problem adaptation |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473224A (en) * | 2013-09-30 | 2013-12-25 | 成都景弘智能科技有限公司 | Problem semantization method based on problem solving process |
| US20150046792A1 (en) * | 2013-08-06 | 2015-02-12 | Educational Testing Service | System and Method for Rendering an Assessment Item |
| CN103250149B (en) * | 2010-12-07 | 2015-11-25 | Sk电信有限公司 | For extracting semantic distance and according to the method for semantic distance to mathematics statement classification and the device for the method from mathematics statement |
| CN105302875A (en) * | 2015-10-09 | 2016-02-03 | 电子科技大学 | Interactive problem solving method, server and system |
| CN105488015A (en) * | 2016-01-22 | 2016-04-13 | 吴以凡 | Test question structuring method |
| CN107562918A (en) * | 2017-09-12 | 2018-01-09 | 北京点易通科技有限公司 | A kind of mathematical problem knowledge point discovery and batch label acquisition method |
-
2018
- 2018-01-29 CN CN201810081517.6A patent/CN110096692B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103250149B (en) * | 2010-12-07 | 2015-11-25 | Sk电信有限公司 | For extracting semantic distance and according to the method for semantic distance to mathematics statement classification and the device for the method from mathematics statement |
| US20150046792A1 (en) * | 2013-08-06 | 2015-02-12 | Educational Testing Service | System and Method for Rendering an Assessment Item |
| CN103473224A (en) * | 2013-09-30 | 2013-12-25 | 成都景弘智能科技有限公司 | Problem semantization method based on problem solving process |
| CN105302875A (en) * | 2015-10-09 | 2016-02-03 | 电子科技大学 | Interactive problem solving method, server and system |
| CN105488015A (en) * | 2016-01-22 | 2016-04-13 | 吴以凡 | Test question structuring method |
| CN107562918A (en) * | 2017-09-12 | 2018-01-09 | 北京点易通科技有限公司 | A kind of mathematical problem knowledge point discovery and batch label acquisition method |
Non-Patent Citations (2)
| Title |
|---|
| 基于本体的语义题库的构建与知识挖掘的研究及应用;李瞿男;《中国优秀硕士学位论文全文数据库 信息科技辑》;20140115(第01期);第69页 * |
| 基于语义题库的智能教辅系统;李幸;《中国优秀硕士学位论文全文数据库 社会科学Ⅱ辑》;20160315(第03期);第22页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110096692A (en) | 2019-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2947581B1 (en) | Interactive searching method and apparatus | |
| US20170330084A1 (en) | Clarification of Submitted Questions in a Question and Answer System | |
| CN111797214A (en) | FAQ database-based problem screening method and device, computer equipment and medium | |
| US20160196490A1 (en) | Method for Recommending Content to Ingest as Corpora Based on Interaction History in Natural Language Question and Answering Systems | |
| WO2020154604A1 (en) | Systems and methods for automating and monitoring software development operations | |
| US9582486B2 (en) | Apparatus and method for classifying and analyzing documents including text | |
| US9734238B2 (en) | Context based passage retreival and scoring in a question answering system | |
| TWI643076B (en) | Financial analysis system and method for unstructured text data | |
| US20160085740A1 (en) | Generating training data for disambiguation | |
| EP3543871A1 (en) | Artificial intelligence-based triple checking method and apparatus, device and storage medium | |
| US20190303522A1 (en) | Document implementation tool for pcb refinement | |
| CN103365974A (en) | Semantic disambiguation method and system based on related words topic | |
| CN114610845A (en) | Multisystem-based intelligent question answering method, device and equipment | |
| CN110096692B (en) | Semantic information processing method and device | |
| TW202123026A (en) | Data archiving method, device, computer device and storage medium | |
| Hulpuș et al. | Knowledge graphs meet moral values | |
| CN111126034B (en) | Medical variable relation processing method and device, computer medium and electronic equipment | |
| Putri et al. | Software feature extraction using infrequent feature extraction | |
| CN110377706B (en) | Search sentence mining method and device based on deep learning | |
| CN109542912B (en) | Interval data storage method, device, server and storage medium | |
| EP4109300A2 (en) | Method and apparatus for querying writing material, electronic device and storage medium | |
| US9946765B2 (en) | Building a domain knowledge and term identity using crowd sourcing | |
| US11068376B2 (en) | Analytics engine selection management | |
| CN115994534A (en) | Government scene hot word mining method, device, equipment and storage medium | |
| US20190303521A1 (en) | Document implementation tool for pcb refinement |
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 |