CN111522960A - A method for constructing water knowledge concept map model - Google Patents

Abstract

The invention discloses a water knowledge concept map model construction method, which integrates and acquires related concept classes and attributes thereof through unstructured data and structured data; carrying out correlation analysis on the concept classes, and constructing a map form of the wading field so as to express the relationship among all concepts or phenomena; forming knowledge reasoning and knowledge discovery through modes of iteration, association, learning and the like, and expanding concepts and attribute contents after consistency detection; and after selecting the actual theme, bringing the theme entity into the map to form the water knowledge concept theme map. Through the mode, the knowledge acquisition and storage method can effectively acquire and store the water knowledge, arrange and induce the related knowledge and discover new knowledge, improves the integrity of wading knowledge, and is used for providing related theories and knowledge induction for theoretical research and engineering practice in wading fields such as hydrology, water resources and hydraulic engineering.

Description

A method for constructing water knowledge concept map model

technical field

The invention relates to the field of knowledge engineering, in particular to a method for constructing a conceptual map model of water knowledge.

Background technique

Knowledge is in line with the direction of civilization, the sum of the results of human exploration of the material world and the spiritual world. With the development of the industry and technology, the knowledge accumulated in the industry has exploded, and it has the characteristics of multi-source, heterogeneous, different words, isolation, confusion, static and so on.

A series of different graphs are stored in graphized knowledge to display the knowledge development process and structural relationship, that is, to use visualization technology to describe knowledge resources and their carriers, to mine, analyze, construct, draw and display knowledge and their interactions. Contacts can be used to solve knowledge storage and application problems.

Accumulating knowledge into a way that is easy for computers to learn and recognize is one of the key technologies of artificial intelligence. It has been widely used in intelligent search, intelligent question answering, personalized recommendation, content distribution and other fields.

In nature, water exists in the ocean, atmosphere, glaciers, underground, inorganic matter, organic matter, organisms, etc. It has the three-property transformation of gas, solid and liquid, and is also the most common additive and product in biology and chemistry. The water environment is the main environment for human survival. The way and process of water flow have a huge impact on the environment, and it is also the focus of the water industry. Therefore, mastering the form of water flow and obtaining the relationship between water wading factors is a powerful supplement to engineering and research in the field of water wading.

SUMMARY OF THE INVENTION

The technical problem to be solved by the present invention is to provide a method for constructing a conceptual map model of water knowledge in view of the defects involved in the background technology, so as to solve the problems of scattered, vague and unguided knowledge in the water-related field.

The present invention adopts the following technical solutions for solving the above-mentioned technical problems:

A method for constructing a water knowledge concept map, comprising the following steps:

Step 1), through unstructured data and structured data, integrate and obtain relevant concept classes and their attributes to form a water knowledge concept class map;

Step 2), according to the water knowledge concept class map and its related subject entities, construct the water knowledge concept subject map in the water wading field, so as to express the relationship between various concepts or phenomena;

Step 3), through iteration, association, learning, etc., to form knowledge reasoning and knowledge discovery, and expand concept and attribute content through consistency detection and knowledge alignment;

Step 4), after selecting the actual topic, bring the topic entity into the graph to form the water knowledge concept topic graph.

As a further optimization scheme of the method for constructing a water knowledge concept map of the present invention, the integration and acquisition of related concept classes and their attributes in step 1) include: "concept class 1-general relationship-concept class 2", "concept class 1-water Related relationship-concept class 2", "concept class-general attribute", "concept class-water-related attribute", "general relationship-general attribute", "water-related relationship-water-related attribute".

As a further optimization scheme of a water knowledge concept map construction method of the present invention, in step 3), consistency detection and knowledge alignment are used to identify whether concepts, attributes, and relationships are repeated or contradictory, including the following steps:

Step 3.1), analyze whether the concept classes with different names in the water knowledge concept class map are the same concept class;

Step 3.2), analyze whether the attributes of different names of concept classes in the water knowledge concept class map refer to the same feature;

Step 3.3), analyze whether the attribute values of the same attribute in the water knowledge concept theme map are the same dimension and can be converted to each other;

Step 3.4), analyze whether there is a contradiction in the relationship.

Compared with the prior art, the present invention adopts the above technical scheme, and has the following technical effects:

The invention is used for the storage and intelligent identification of knowledge in the hydrological field, can solve the problems of knowledge dispersion, ambiguity, and no guidance in the water-related field, and has the effect of merging, summarizing and arranging knowledge and visualizing it.

Description of drawings

Fig. 1 is a schematic diagram of the construction and application flow of a knowledge concept map of the present invention;

Fig. 2 is the schematic diagram of water knowledge concept class diagram of the present invention;

3 is a schematic diagram of the concept class-attribute of surface runoff in the concept class map of water knowledge according to the present invention;

Figure 4 is a schematic diagram of the theme map of the water knowledge concept in the Yalong River Basin;

Figure 5 is a schematic diagram of the entity-attribute-attribute value of the two estuary stations in the conceptual theme map of water knowledge in the Yalong River Basin.

Detailed ways

Below in conjunction with accompanying drawing, the technical scheme of the present invention is described in further detail:

The present invention may be embodied in many different forms and should not be considered limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. In the drawings, components are exaggerated for clarity.

As shown in Figure 1, the present invention discloses a method for constructing a water knowledge concept map model, including:

Through the unstructured data and structured data related to water, integrate and extract consistent common concept classes, attributes, and relationships, and establish the concepts of "Concept Class 1-General Relationship-Concept Class 2", "Concept Class 1-Water Related" Relation-Concept Class 2", "Concept Class-General Attribute", "Concept Class-Water-Related Attribute", "General Relationship-General Attribute", "Water-Related Relationship-Water-Related Attribute" water knowledge concept class storage system, get and represent water-related knowledge concepts to form a water knowledge concept class map; and according to the selected topic and water knowledge concept class map, obtain relevant subject entities, and perform "class-entity" correspondence, so as to obtain "entity 1-general relationship-entity" 2", "entity 1-water related relationship-entity 2", "entity-general attribute (-attribute value)", "entity-water-related attribute (-attribute value)", "entity-special attribute (-attribute value)" ”, “general relation-general attribute (-attribute value)”, “water-related relation-water-related attribute (-attribute value)” of water knowledge concept theme map; wherein, attribute value can be a number or character with special meaning, You can also continue to build relationships and attributes for concept classes.

Through iteration, association, learning, etc., the knowledge evaluation content is mined to form knowledge reasoning and knowledge discovery. After consistency detection, the concept map of water knowledge is expanded and improved. Among them, consistency detection and knowledge alignment are to identify whether concepts, attributes, and relationships are repeated or contradictory, including but not limited to: a), analyzing whether concept classes with different names in the water knowledge concept class map are the same concept class; b), Analyze whether attributes of different names of concept classes in the water knowledge concept class map refer to the same feature; c), analyze whether the attribute values of the same attribute in the water knowledge concept theme map are the same dimension and can be converted to each other; d), analyze Whether there is conflict in the relationship.

Further, the sources of water-related knowledge acquisition include but are not limited to:

Involved in hydrology and water resources, meteorology, geographic science, water conservancy and hydropower engineering, water engineering and other related industries, including but not limited to empirical formulas, physical formulas and chemical formulas, professional books and online texts, videos (images), audio, social Data such as water cycle information, network public opinion, etc.; and,

Expert system analysis and expert experience analysis involved in hydrology and water resources, meteorology, geographic science, water conservancy and hydropower engineering, water engineering and other related industries; and,

Public domain knowledge graphs, including but not limited to general knowledge graphs such as Google Knowledge Graph, Baidu Knowledge Graph, and Sogou Knowledge Cube.

Embodiments of the present invention include:

Establish various water-related concepts in the hydrological cycle and their relationship with each other, and form a basic concept map of the hydrological cycle. Among them, the rounded rectangular box represents "water knowledge concept class"; "solid line-hollow oval box-solid line" represents "general relationship", the arrow of the solid line represents the direction of the relationship, and no arrow is a parallel relationship; "dotted line- "Grey oval box-dotted line" represents "water related relationship"; "dotted line-hollow rectangle box" represents "general property"; "dotted line-gray rectangular box" represents "water related property".

With reference to the experience of hydrology experts and the content of the "Hydrology Handbook", you can extract watersheds, sub-basins, soils, river channels, surface runoff, mountains, plains, basins, underground runoff, plants, snow, rain, clouds, wind, oceans, reservoirs, lakes , measuring station, hydrological station, rainfall station and other water-related concept categories, and generate the general relationship and water-related relationship between each other, forming Figure 2 "Schematic Diagram of Water Knowledge Concept Category".

Through expert experience, there is no duplication or conflict between these concepts, relationships and attributes, which conforms to consistency detection and knowledge alignment;

Each concept class and some relationships have attributes, forming a “concept class (relationship)-attribute” graph, as shown in Figure 3, “Water Knowledge Concept Class Graph Surface Runoff Concept Class Diagram”, which can be used as an extension of Figure 2.

According to the experience of experts, the concept class "watershed" has general attributes such as "name", "range", "altitude", and "area"; the concept class "hydrological station" has "name", "longitude", "latitude", General attributes such as "elevation", and water-related attributes such as "observation items"; the concept class "surface runoff" has "name", "estuary", "flow through province", "transboundary type", "river type", "superior General attributes such as "river", "river length", "watershed area", "code", and "station", there are "multi-year average runoff depth", "multi-year average precipitation depth", "average gradient", "flood", " Water-related attributes such as "Heyuan"; the relationship "Sag" has general attributes such as "Sag Area" and "Sag Origin".

Record "Concept Class 1-Relation-Concept Class 2" as in Table 1; record "Concept Class (Relation)-Attribute" as in Table 2.

Table 1 Concept Class - Relationship Table

Table 2 Concept Class (Relationship) - Attribute Table

Discover new concept classes through relationships, expand the attributes of the concept classes in Table 1, and update Figure 2 and Figure 3;

Taking the Yalong River Basin as the theme, complete the Yalong River Basin Water Knowledge Concept Theme Map Case:

Corresponding to the "sub-watershed-Yalong River Basin", obtain the concept categories "soil" and "watershed" related to the "sub-basin", and establish "soil-red soil", "soil-red soil", "soil-cinnamon soil" ”, the class-entity correspondence of “Watershed-Yangtze River Basin”;

Continuing through the relationship, more conceptual entities related to water knowledge in the Yalong River Basin can be obtained, and a schematic diagram of the conceptual theme map of water knowledge in the Yalong River Basin is formed in Figure 4. The “class-entity” situation is recorded in Table 3.

Table 3 "Class-entity" table of water knowledge topics in the Yalong River Basin

Entity/Relation Number type Type number entity/relationship name Relationship Notes 1 entity 2 Yalong River Basin 2 entity 1 Yangtze River Basin 3 entity 3 red soil 4 entity 3 red soil 5 entity 3 cinnamon 6 entity 10 Lianghekou Station 7 relation 5 Confluence … … …

Determine "entity-attribute-attribute value" and "relationship-attribute-attribute value" according to "class-entity" and "class-attribute". Through association, the subject of Yalong River Basin has entity name "Two River Mouth Station", which belongs to the concept class "Hydrological Station", generates "entity-attribute-attribute value" and stores it in Table 4, and generates Yalong as shown in Figure 5 The schematic diagram of the two estuary stations in the concept map of water knowledge in the river basin, in which the borderless attribute value is a number or character with special meaning. Obtain new attributes "province" and "feature element", and after "knowledge discovery", determine "province" as a new general attribute, in Table 4 the attribute number starts with "n", add an item to Table 2, and update the map ; "Characteristic element" is a special attribute of the entity "Lianghekou Station". In Table 4, the attribute number starts with "s", and it is represented by a dotted box in Figure 5.

Table 4 "entity (relation)-attribute-attribute value" table of water knowledge concept theme map in the Yalong River Basin

Entity/Relation Number property number attribute value New/Special Properties 6 1 Lianghekou Station 2 101.01E 3 30.19N 4 2780.5 4 Datum 5 rainfall 5 water quality 5 run-off n1 Sichuan province s1 flood characteristic element s1 precipitation characteristic element

Through the water knowledge concept theme map in Figure 4 and Figure 5, relevant knowledge about the theme "Yalong River Basin" can be obtained, which is convenient for selecting relevant data and data for service application.

The present invention first establishes a water knowledge concept map, including knowledge point decomposition, class extraction, attribute extraction, relationship extraction, consistency detection, knowledge alignment and establishment of a water knowledge concept class map, wherein:

The water knowledge concept map includes the water knowledge concept class map and the water knowledge concept theme map. The water knowledge concept class map is an integration of the abstract concepts of water knowledge, excluding specific entity content; The integration of thematic knowledge entity-relationship-attribute-attribute value generated by the correspondence between the entity concept and the water knowledge concept class map;

The relationship in the concept map of water knowledge is divided into general relationship and water-related relationship. General relationship represents the existing relationship between two or more concept classes, mainly subject-predicate-object relationship, parallel relationship, inheritance relationship and related relationship, etc.; The water-related relationship is the change of water between different concept classes, the way of existence and the influence on the concept class;

The attributes of the water knowledge concept map are divided into general attributes, special attributes and water-related attributes. The general attributes are the attributes that each water knowledge concept class must contain; the special attributes are the subject entities corresponding to the concept classes but not the water knowledge concept classes. The attributes of water-related attributes are attributes brought about by the existence and changes of water in the knowledge concept category;

Then perform knowledge discovery and graph update, including knowledge evaluation, knowledge reasoning, and knowledge discovery;

Finally, the application of water knowledge concept map is established, including topic selection, class-entity correspondence, attribute value acquisition, establishment of water knowledge concept theme map and service application.

It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless defined as herein, are not to be taken in an idealized or overly formal sense. explain.

The specific embodiments described above further describe the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above descriptions are only specific embodiments of the present invention, and are not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (3)

1. A water knowledge concept map construction method is characterized by comprising the following steps:

step 1), integrating and acquiring related concept classes and attributes thereof through unstructured data and structured data to form a water knowledge concept class map;

step 2), constructing a water knowledge concept theme map in the wading field according to the water knowledge concept class map and relevant theme entities thereof so as to express the relationship among all concepts or phenomena;

step 3), forming knowledge reasoning and knowledge discovery through modes of iteration, association, learning and the like, and expanding concepts and attribute contents through consistency detection and knowledge alignment;

and 4) after selecting the actual theme, bringing the theme entity into the map to form the water knowledge concept theme map.

2. The method for constructing a water knowledge concept graph according to claim 1, wherein the step 1) of integrating and acquiring related concept classes and attributes thereof comprises: "concept class 1-general relationship-concept class 2", "concept class 1-water correlation relationship-concept class 2", "concept class-general attribute", "concept class-water correlation attribute", "general relationship-general attribute", "water correlation relationship-water correlation attribute".

3. The method for constructing the water knowledge concept graph according to claim 1, wherein consistency detection and knowledge alignment in the step 3) are used for identifying whether the concepts, attributes and relationships are repeated or inconsistent, and the method comprises the following steps:

step 3.1), analyzing whether the concept classes with different names in the water knowledge concept class atlas are the same concept class or not;

step 3.2), analyzing whether the attributes of different names of the concept classes in the water knowledge concept class atlas refer to the same characteristic or not;

step 3.3), analyzing whether attribute values of the same attribute in the water knowledge concept topic map are in the same dimension and can be mutually converted;

and 3.4) analyzing whether the relationship is contradictory.

Images