CN105718649A - Construction method for paper book capable of hierarchically displaying knowledge network - Google Patents

Abstract

The present invention provides a paper book construction method for hierarchically displaying knowledge networks, including: constructing knowledge point objects: decomposing the field knowledge involved in paper books into a series of knowledge points without name conflicts; constructing paper book catalogs : According to the theme of the paper book, all the knowledge point objects are organized according to the causal relationship using the depth-first search method to form a knowledge trunk tree, and displayed by the directory structure display method; according to the obtained directory tree, from the beginning of the directory, the corresponding The knowledge point description of the entry is written into the content of the paper book body with page numbers sequentially marked. At the same time, the beginning page number of each knowledge point and some symbols representing the relationship between related entries of the directory tree are also recorded in the text; constructing knowledge points and Other search methods. The above method can make the compiled paper book more clear in hierarchy, more convenient to use, and more conducive to establishing the internal relationship between the internal knowledge of the paper book.

Description

A paper book construction method for hierarchically displaying knowledge network

technical field

The invention belongs to the field of book publishing, and more specifically relates to a paper book construction method for hierarchically displaying knowledge networks, which can organize book contents with complex network relationships in a hierarchical association and fast knowledge point query manner.

Background technique

Knowledge is people's understanding of the objective world. Knowledge generally exists in the form of a knowledge network, that is, knowledge can be abstracted into several knowledge points, and then connected to form a knowledge network.

Paper books, that is, paper books, are important media for human knowledge and civilization inheritance and dissemination. Essentially, a paper book is a one-dimensional or two-dimensional description of a knowledge network. The construction method of paper books is: use natural language or other languages to describe knowledge and ideas, and then give a main directory according to a certain relationship to organize them, and provide indexes and other retrieval methods.

With the development and progress of computer technology, people often use electronic technology products to record knowledge and ideas, thus producing e-books. In a broad sense, any electronic technology product is a reflection of knowledge, and can also be considered as a record and application of knowledge and ideas, but the e-book here is narrowly defined, referring to products that can be read and acquired in a computer. Its main purpose is the learning of knowledge and the understanding of ideas.

Because e-books have a strong connection ability to knowledge points, e-books can help readers build knowledge networks by establishing connections between knowledge points, so as to effectively master relevant knowledge. Therefore, the use of e-books is the development trend of book publishing.

However, paper books with a history of thousands of years will still have value for a long time, mainly because:

(1) Paper books have a strong sense of existence, intimacy, naturalness, and randomness. People can touch it when they are reading, and they will have a strong psychological feeling, and they can't put it down.

(2) Paper books have a strong sense of integrity and detail. It is difficult for an e-book to produce a sense of the whole, although it has a powerful zoom-in and zoom-out function, but if the magnification factor is too high, it is not easy to form an overall concept, and if the magnification is too high, it is difficult to see the subtle parts.

(3) Paper books are easy to pass on. Paper books record knowledge directly with text symbols. As long as they understand the meaning of text symbols, people can recognize them no matter how old they are, so we can read books from ancient times. The e-book is not the case, it needs an interpreter, that is, an interpreter is needed to read in electronic data and convert it into visual text symbols. If the interpreter of a certain e-book does not exist, then the e-book cannot be used.

(4) Paper books are convenient for free-form knowledge learning, innovation and creation. The functions of e-books are very powerful, but special tools and operation methods are required. For example, the required knowledge points can be quickly found through keyword search. However, when people carry out knowledge learning and innovation, a more free and casual way may be more beneficial. For example, it may be more valuable for the learning and innovation of knowledge to write down the learning experience somewhere in the paper book.

Whether it is a traditional book or an e-book, its purpose is to express intricate knowledge and ideas, which is convenient for users to learn and apply. Obviously, the clearer and more convenient it is to express this complex relationship between knowledge and thought, the better the learning and application effect of knowledge will be. Therefore, drawing lessons from the characteristics of e-books to express knowledge relations, designing a paper book construction method that more clearly expresses knowledge relations is very valuable for knowledge learning, application and innovation.

Contents of the invention

In order to enable paper books to better represent knowledge relationships, that is, represent knowledge networks, it is necessary to apply some theories and methods related to knowledge representation in the field of artificial intelligence research.

There are many representation methods of knowledge, such as state space method, predicate logic, ontology, semantic network, frame system, production representation, Petri net representation, neural network and other methods. These methods can be attributed to the standard model of knowledge representation, and finally form a knowledge network. According to the standard model of knowledge representation, knowledge is a reflection of the objective world, and because the objective world is multidimensional, knowledge is also multidimensional. Knowledge application is inseparable from knowledge network, so knowledge points are interrelated. To effectively represent knowledge, we must pay attention to two points: one is to represent knowledge from multiple dimensions, and the other is to establish the relationship between knowledge.

Obviously, e-books have strong advantages for knowledge representation. Therefore, we first carefully analyze the construction method of the e-book, and then describe how to construct the paper book.

If the knowledge has the same knowledge granule size, or its knowledge granules are indivisible in an application, knowledge representation can be performed, and it can be reasoned and solved. This is the basic guarantee for the success of many artificial intelligence systems. But for an application field, it is impossible for people to build such a knowledge system composed of indivisible knowledge points, because this not only makes the knowledge system very large, but also greatly reduces the solution efficiency of the knowledge system. Therefore, the construction of e-books cannot use general knowledge representation methods, but only knowledge representation methods such as expert systems. The present invention adopts the knowledge representation method of problem orientation analysis and decision (POAD, Problem Oriented Analysis and Decision) to carry out the knowledge construction of e-book, promptly uses the expert system of problem orientation analysis and decision expert system (POADES, Problem Oriented Analysis and Decision Expert System) to carry out knowledge representation. The basic concepts of expert systems for problem analysis and decision-making (POAD) are as follows:

(1) Knowledge dictionary

The knowledge dictionary is used to describe the objects in the knowledge system, and mainly includes three aspects: object name, object type and object description. Therefore, the knowledge dictionary D can be represented by triples as follows:

D=(O,T,E)

Here O represents the object, T represents the classification of the object, which is commonly referred to as the part of speech, and E represents the semantic interpretation of the object O.

In POAD, there are 5 commonly used objects: subject object, problem object, problem phenomenon object, principle knowledge object and process knowledge object, represented by O _M , _OPB , OP _PH , O _PL , O _PR respectively. The objects are collectively referred to as MP4 objects.

(2) POAD knowledge unit

The POAD expert system is a problem-centered system that complies with human thinking habits to construct and apply knowledge. It adopts a knowledge representation that supports problem analysis and decision-making and is centered on problem objects. Its knowledge units, that is, knowledge points, can be shown in Figure 1. express.

The knowledge system K can be described as follows:

K＝(O _M , O _PB , O _PH , O _PL , O _PR , R)

Here R represents the relationship between various objects, namely:

R=(O ₁ ,O ₂ )

Here O ₁ , O ₂ represent any two objects.

relationship between objects

In POAD, the relationship between two objects is divided into two types: simple correlation relationship and causal relationship.

The simple correlation R _S is defined as follows:

R _S =(O _PB ,O ₂ )

In the formula, _O2 represents other objects other than _OPB type. R _S is represented by a single line in Figure 1.

The causality R _C is defined as follows:

R _C ＝(O _PB1 ,O _PB2 )

A causal relationship represents a relationship between problem objects. Because trees are often used to describe causal relationships, they are called causal trees. The causal tree T can be represented by the following formula:

T＝(O _PB , R _C )

R _C is represented by a double line in Figure 1. Generally speaking, the relationship represented by R _C has two types: "and" and "or". The lines with arrows in Figure 2 indicate the relationship of "or", while the line segments connected by arcs indicate the relationship of "and". For example, problem A is jointly determined by B, C, and D, and as long as one of H, I, and J occurs, C must occur.

Causality is people's understanding of the fundamental principles of nature, reflecting the decomposability of problem objects. The termination condition of causality decomposition is application-dependent.

POAD is consistent with the thinking habits of human beings. The expert system POADES (Problem Oriented Analysis and Decision Expert System) developed by POAD has the advantages of easy development, unlimited knowledge, and convenient reasoning. Construct.

Since paper books record knowledge on static paper, they must be limited to a two-dimensional paper plane. From a formal point of view, paper books record knowledge in a two-dimensional manner. From a local point of view, each knowledge point in the book itself can be multi-dimensional. For example, mathematical expressions can be regarded as expressing knowledge in a multi-dimensional way. Books actually record knowledge in a one-dimensional way. Therefore, one difference between paper books and e-books is that paper books cannot directly realize the jump of knowledge points through link relationships like e-books. A certain jump relationship must be set to make paper books express complex knowledge. relation. Another difference between paper books and e-books is that paper books themselves are static, and it is impossible to automatically reason and solve knowledge when readers use them, so some simplifications can be made.

In order to make paper books better express knowledge and ideas, the present invention proposes a paper book construction method based on a problem-oriented analysis and decision (POAD, Problem Oriented Analysis and Decision) expert system. This method firstly analyzes and synthesizes the content to be expressed in the paper book, and divides the knowledge units according to the main objectives and core issues of the paper book, and establishes the knowledge point object and its description. For paper books that need to express domain knowledge very accurately, MP4 objects can be established on the basis of analysis, induction, and synthesis of the knowledge and ideas to be expressed, and appropriate knowledge particles can be selected for each object. For paper books that are generally required, it can be simplified. In the present invention, the knowledge point object is used to replace the MP4 object, that is, only the knowledge point object is established. This object is an object with the problem object as the core, which simplifies the entire knowledge unit into one object, and does not discuss separately the subject object, phenomenon object, principle knowledge object and process knowledge object related to the problem object. The content of the MP4 object is included in the knowledge point object, which requires readers to understand through reading. Each knowledge point object contains two parts, one part is the description of the knowledge point object at the current level, and the other part is the internal composition of the knowledge point object, which is represented by its sub-knowledge point object.

Then build a knowledge trunk tree according to the main goal of the paper book, and connect all the knowledge points into a knowledge trunk tree by using the depth-first search method according to the causal relationship. On the knowledge trunk tree, each knowledge point is organized according to the parent-child relationship. This parent-child relationship is mainly a classification relationship and a composition relationship, and it can also be other relationships. What they form is an AND-OR tree. Such trees look like trees, but are essentially graphs. In addition, any two nodes on the tree are allowed to be connected, that is, a node can have a relationship with a node on another branch, thus forming a knowledge network. This kind of knowledge trunk tree constitutes the catalog of paper books, which can not only reflect the main idea of paper books, but also can more deeply and comprehensively reflect the internal relationship of paper book knowledge through the connection relationship on the trunk tree.

Finally, it provides functions such as regular retrieval and knowledge point retrieval for the contents of paper books.

In the following discussion, in order to be consistent with conventional books, the knowledge trunk tree is called a directory, and the nodes on it are entries, which represent knowledge points. Knowledge points and their names are unique in paper books, but entries with the same name are not unique, that is, entries with the same name can exist in the catalog, and they all point to the same knowledge point. The method comprises the steps of:

1) Build knowledge point objects

Decompose the domain knowledge involved in paper books into a series of knowledge points without name conflicts;

For each knowledge point, it is distinguished by several independent text symbols (may include graphic symbols, mathematical symbols, etc.), that is, each knowledge point is represented by a knowledge point name and its corresponding symbol set. The knowledge point only contains the description of the current layer, and the sub-knowledge point only lists the entries without displaying the content.

2) Construct a catalog of paper books

According to the theme of the paper book, all knowledge point objects are organized according to the causal relationship using the depth-first search method to form a knowledge trunk tree, and displayed by the directory structure display method;

Its main steps are as follows:

(2.1) Set the use flags of all knowledge points to "unused", set the general title of the paper book, that is, the total knowledge points, as the current entry, record the current entry on the directory tree, and set the current entry and the corresponding knowledge points The sign of use is "use";

(2.2) Find all the sub-knowledge points related to the current entry, that is, the current knowledge point, as the sub-entries of the current entry, and arrange them in a certain order;

(2.3) For all sub-entries of the current entry, select each sub-entry according to the set arrangement order, and record it on the directory tree;

(2.4) If the use flag of the knowledge point is "use", then set the use flag on the sub-entry to "reference", otherwise set its use flag to "use";

(2.5) If there is a sub-entry in this sub-entry and the use mark of its corresponding knowledge point is "unused", then set the use mark of the corresponding knowledge point to "use" and repeat (2.2)-(2.5) until all knowledge points The dots are all used, and you end up with a directory tree.

3) Construct the text of the paper book

According to the catalog tree obtained in 2), from the beginning of the catalog, write the knowledge point description of the corresponding entry into the content of the paper book text with page numbers sequentially marked, and record the start page number of each knowledge point in the text and some symbols representing the relationship between related entries in the directory tree, the steps are as follows:

(3.1) Obtain an entry from the directory tree, set it as the current entry, and find the corresponding knowledge point.

(3.2) If the use flag of the current entry is "use", add the description of the current knowledge point at the end of the current text page of the paper book, and record the starting position of the knowledge point at the page number of the current entry. If the current entry If the usage flag is "reference", then find the page number of the start position of the corresponding knowledge point, and write the page number to the page number of the current catalog, and add "*" in front of the referenced page number to distinguish it.

(3.3) On the first page of the current knowledge point, if the previous entry of the current entry is a parent node, mark the parent-child relationship symbol "┕" between the entry and the previous entry at the beginning or the left margin, and do not mark Indicates that the relationship between the previous entry and the current entry is a sibling relationship.

(3.4) In each page of the text where the current knowledge point is located, if the knowledge point is at the forefront of the current page, mark the starting page number of the knowledge point at the header of the current page.

(3.5) On the last page of the current knowledge point, if the current entry is a leaf node, mark the terminator symbol "■" of the knowledge point at the end or in the margin on the right. Those not marked are branch nodes, indicating the following part It is a child node of the current knowledge point.

4) Build retrieval methods for knowledge points and other content

The steps are as follows:

(4.1) List the names of all knowledge points and their page numbers at the back of the paper book in a certain order.

(4.2) List other keywords and their page numbers in all paper books, such as all conceptual terms and their page numbers, in a certain order at the back of the paper book.

The beneficial effect brought by the present invention is that the proposed method can make the compiled paper books clearer, more convenient to use, and more conducive to establishing the internal relationship between the internal knowledge of paper books, thereby helping authors to accurately Accurately organize and express the knowledge and ideas that need to be expressed, so that readers can more easily and deeply understand the knowledge content and complex relationships of paper books through reading and reviewing the directory structure and text, and at the same time enable readers to still retain the traditional reading tradition. The fun and variety of paper books.

Description of drawings

Figure 1 is a structural diagram of knowledge units in POAD;

Figure 2 is a causal tree diagram in POAD;

Fig. 3 is a catalog structure diagram of a paper book;

Fig. 4 is a text structure diagram of a paper book;

Fig. 5 is a flow chart of paper book catalog construction;

Fig. 6 is a flow chart of paper book text construction.

detailed description

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not constitute a conflict with each other.

In the present invention, it is assumed that the task of constructing a paper book is to organize these knowledge into a paper book capable of expressing a certain knowledge network relationship for readers to read under the existing knowledge level and related materials.

In the following discussion, the knowledge point is the construction unit of the paper book, and the catalog is the knowledge organization structure of the paper book, that is, the mapping of the knowledge trunk tree of the book, and each entry in the catalog corresponds to the knowledge trunk tree. A node also corresponds to a knowledge point in the text of the paper book, and each knowledge point in the text corresponds to at least one entry in the table of contents. The method comprises the steps of:

1) Build knowledge point objects

Decompose the domain knowledge involved in paper books into a series of knowledge points without name conflicts;

For each knowledge point, it is distinguished by several independent text symbols (may include graphic symbols, mathematical symbols, etc.), that is, each knowledge point is represented by a knowledge point name and its corresponding symbol set. The knowledge point only contains the description of the current layer, and the sub-knowledge point only lists the entries without displaying the content.

The most basic work of constructing a paper book is to sort out the knowledge and ideas expressed by the paper book, and decompose them into a series of knowledge point objects without name conflicts. The setting of the knowledge particle size of knowledge points is very important. Obviously, paper books cannot be made into a very complete knowledge system, that is, the knowledge particles of the knowledge points involved cannot be so detailed that they cannot be divided. This is because on the one hand, it is very difficult to achieve On the other hand, even if all knowledge points can be divided into indivisible knowledge particles, there is no way to record them in paper books, because the number of knowledge points will be extremely large. In fact, even the author himself does not necessarily have a clear understanding of all the knowledge involved in the book, so there is a certain amount of ambiguity in any paper book. Generally speaking, the setting of knowledge points should follow the following principles: it can reflect the mainstream knowledge in the field; it can solve related problems; the common sense knowledge of readers in this field can be brief; it can meet the readers' usage habits, etc.

Knowledge point objects must satisfy the hierarchical relationship. The relationship between knowledge point objects is complex, but we can use the POAD method to stratify them according to a certain topic, that is, all knowledge points establish parent-child relationships based on generalized causal relationships, so that a knowledge trunk tree can be formed. This kind of knowledge trunk tree is actually an AND or graph, which can represent the relationship between more complex knowledge points. All parent-child relationships can be simply divided into two categories, one is classification relationship and the other is composition relationship. If it is a classification relationship, the relationship between each child node is an OR relationship, and if it is a composition relationship, then the relationship between each child node is an AND relationship.

Through the AND or graph, multiple knowledge points in the field can be naturally connected together, but if each knowledge point is expanded, there will be a lot of repeated parts. In order to solve this problem, it is necessary to allow two knowledge points that do not have a parent-child relationship to establish a related relationship, that is, nodes on multiple knowledge trunk trees can refer to the same knowledge point, so the knowledge trunk tree is actually called a knowledge trunk graph.

Therefore, the knowledge points in the paper book can be represented by the names and description symbols of the knowledge points, which will be inserted into the text of the paper book in the following steps, and the knowledge trunk tree corresponds to the catalog on the paper book, each A node corresponds to an entry on a directory, where multiple entries may be allowed to be identical.

2) Construct a catalog of paper books

According to the theme of the paper book, the process of organizing all the knowledge point objects according to the causal relationship using the depth-first search method to form a knowledge trunk tree, and displaying it with the catalog structure display method is called the construction process of the paper book catalog.

The core of any kind of book is the table of contents, which essentially represents the knowledge system of the entire book. In general paper books, although the catalog is also constructed according to certain rules, such as classification relationship and composition relationship, in order to reduce the catalog level, eliminate redundancy, facilitate reading, highlight key points, etc., in fact, many paper books Directories are structured in a mixed fashion, and even some logical confusion. The main function of this kind of catalog is not to organize knowledge, but to facilitate the query and application of the content in the book. Knowledge points and their interrelationships need to be constructed by readers themselves.

The defect of the traditional paper book is caused by the fact that the book is a two-dimensional medium. Obviously, if e-books are used, the association between knowledge points can be realized by establishing the link relationship between knowledge points, so that the relationship between knowledge points can be pre-established in e-books, which is very important for the learning and application of knowledge. very important. Although paper books do not have the inherent advantages of e-books, by marking some information on paper books, it can also have a certain knowledge connection function.

The method for establishing the association between knowledge points in the present invention is: establishing a catalog and marking the page number of the knowledge point on the catalog entry. When creating the catalog, each knowledge point must comply with the regulations made in step 1), that is, they must meet the specified parent-child relationship. At the same time, in order to make it easier for readers to use paper books, the catalog must be arranged in the order of depth priority, so when When readers refer to a knowledge point, the knowledge point must be in the front part of the current knowledge point text, that is to say, it must be a knowledge point that has already been learned.

In the present invention, there are only two kinds of parent-child relationships in the directory: classification relationship and composition relationship, that is, the directory is an AND or graph. Because these two relationships are very easy to distinguish, no special marking is made in the table of contents, that is, the AND or relationship needs to be determined by the reader when reading.

The main steps to build a paper book catalog are as follows:

(2.1) Set the use flags of all knowledge points to "unused", set the general title of the paper book, that is, the total knowledge points, as the current entry, record the current entry on the directory tree, and set the current entry and the corresponding knowledge points The sign of use is "use";

(2.2) Find all the sub-knowledge points related to the current entry, that is, the current knowledge point, as the sub-entries of the current entry, and arrange them in a certain order;

(2.3) For all sub-entries of the current entry, select each sub-entry according to the set arrangement order, and record it on the directory tree;

(2.4) If the use flag of the knowledge point is "use", then set the use flag on the sub-entry to "reference", otherwise set its use flag to "use";

(2.5) If there is a sub-entry in this sub-entry and the use mark of its corresponding knowledge point is "unused", then set the use mark of the corresponding knowledge point to "use" and repeat (2.2)-(2.5) until all knowledge points The dots are all used, and you end up with a directory tree.

3) Construct the text of the paper book

The directory tree obtained in step 2) records the hierarchical relationship between the knowledge points, but does not record the specific content of the knowledge points. The text of the paper book is used to record all knowledge point content information. The production of the text of the paper book needs to meet the following requirements:

A) Completely record the content of knowledge points

The content of a knowledge point includes the name or entry of the knowledge point, the description of the knowledge point, and the sub-knowledge points of the knowledge point, that is, in the text, any complete record of a knowledge point must include the entry of the knowledge point, the Description and the sub-entries of this knowledge point and their descriptions, the entry of this knowledge point is essential, and the complete record of any knowledge point only appears once, and the others can only be references.

B) Store the starting position of the current knowledge point into the corresponding entry in the directory

Each knowledge point has only one description in the text, so there is only one starting position for it. However, there may be multiple entries in the directory that use or reference this knowledge point, so it is necessary to ensure that these entries all point to the beginning of the same knowledge point. However, since the catalog must provide readers with overall orientation, it is necessary to mark the reference entry, which not only distinguishes whether the entry is the first use of the knowledge point, but also does not affect the overall orientation of the catalog. References in a knowledge point are part of the description of the knowledge point and cannot constitute a knowledge point alone. It is equivalent to the general description text in the knowledge point, except that the corresponding knowledge point start position page can be attached to these entries.

C) Indicate the starting position of the first knowledge point on the current page

There may be multiple knowledge points on the current page, but except for the first knowledge point on the page, the starting positions of other knowledge points are all on this page, so there is no need for special marking, and the first knowledge point may be in the previous text page , or in the text page of the current page (at this time, the first line of the text is the starting position of the knowledge point), at this time, it needs to be marked at the header to facilitate readers' inquiries.

D) Distinguish the relationship between the current node and the previous node

Each entry in the current text page, that is, there may be three types of relationships between the current node and the previous node, one is the parent-child relationship, the other is the brother relationship, and the other is the relationship between the beginning of all text pages, that is, the current entry is the entire page The start entry of the quality book. Obviously, the start item does not need special attention, so only parent-child relationship and brother relationship are distinguished, because the distinction between these two types of relationships is beneficial to the understanding of the content and relationship between knowledge points.

E) Distinguish the relationship between the current node and the next node

The end of each knowledge point in the current text page means the end of the current knowledge point, and there may be three types of relationships between the corresponding entries, that is, the current node and the next node, one is parent-child relationship, and the other is brother The other type is the end relationship of all knowledge points, that is, there is no longer any knowledge point content behind. Therefore, the current node can be divided into two types: leaf node and branch node. Distinguishing these two types of relationships is also beneficial to the understanding of the content and relationships between knowledge points.

The steps to construct the body of a paper book are as follows:

(3.1) Obtain an entry from the directory tree, set it as the current entry, and find the corresponding knowledge point.

(3.2) If the use flag of the current entry is "use", add the description of the current knowledge point at the end of the current text page of the paper book, and record the starting position of the knowledge point at the page number of the current entry. If the current entry If the usage flag is "reference", then find the page number of the start position of the corresponding knowledge point, and write the page number to the page number of the current catalog, and add "*" in front of the referenced page number to distinguish it.

(3.3) On the first page of the current knowledge point, if the previous entry of the current entry is a parent node, mark the parent-child relationship symbol "┕" between this entry and the previous entry at the beginning or left margin, and do not mark Indicates that the relationship between the previous entry and the current entry is a sibling relationship.

(3.4) In each page of the text where the current knowledge point is located, if the knowledge point is at the forefront of the current page, mark the starting page number of the knowledge point at the header of the current page.

(3.5) On the last page of the current knowledge point, if the current entry is a leaf node, mark the terminator symbol "■" of the knowledge point at the end or in the margin on the right. Those not marked are branch nodes, indicating the following part It is a child node of the current knowledge point.

4) Build an index for knowledge points and other content retrieval

In many cases, searching based on the names of knowledge points or keywords has high practical value, so conventional searching is of great significance. The steps to create an index are as follows:

(4.1) List the names of all knowledge points and their page numbers at the back of the paper book in a certain order.

(4.2) List other keywords and their page numbers in all paper books, such as all conceptual terms and their page numbers, in a certain order at the back of the paper book.

The following takes the construction process of a paper book as an example to illustrate the implementation process of the paper book construction method of the present invention. The present invention adopts the computer system Office software as the means of realization, and takes the accompanying drawings shown in Figure 3 to Figure 6 as the object of description, and the process and implementation plan of the paper book construction method disclosed by the present invention are described as follows:

Step 1: Build knowledge point objects

Decompose the domain knowledge involved in the paper book into a series of knowledge points without name conflicts. For each knowledge point, it is stored in a Word file, wherein the file name of the file is named after the knowledge point, and the content of the file is distinguished by several independent text symbols (may include graphic symbols, mathematical symbols, etc.). If the knowledge point contains sub-knowledge points, the entry names of the sub-knowledge points can be added to the file, and the content of the sub-entries can be selected according to the needs when constructing the text of the paper book.

Step 2: Build a physical book catalog

Generate a file for recording the table of contents of the current paper book. Use the title of the current paper book as the root node of the directory, but do not write it in the directory. Then start from the root node, search all knowledge points according to the depth-first principle, and construct all directory entries. Its main steps are shown in Figure 5:

Step 501: start.

Step 502: Set the start entry as the current entry. Create a knowledge point processing record file with the name: Knowledge.xls. The file has 3 columns, which are: knowledge point name, usage flag, and starting page number. Fill in the name of each knowledge point, and set the use flag to "unused". Create a directory file: mulu.doc to record the directory of the paper book. As shown in Figure 3, each line of record in the file mulu.doc represents an entry, and each entry includes two parts, one part is the name of the entry, and the other part is the page number. Each entry is set to a corresponding format according to its level, for example, a first-level entry is set to a first-level title format, and so on.

Step 502 is only for the root node, and does not need to write the current entry in the directory.

Step 503: Find all sub-entries of the current entry, and record them in order in another text file named mulu.tmp.

Step 504: Check whether there are unused entries in the file mulu.tmp, if yes, go to step 505, otherwise go to step 511.

Step 505: Take an entry from mulu.tmp and set it as the current entry, write it in the directory file mulu.doc, and then delete the entry from mulu.tmp.

Step 506: Find the record of the corresponding knowledge point in the file Knowledge.xls according to the item name.

Step 507: If the use flag of the knowledge point is "used", go to step 512, otherwise go to step 508.

Step 508: Write the flag "use" at the page number of the current entry in the mulu.doc file.

Step 509: If the current entry has sub-entries, go to step 510, otherwise go to step 504.

Step 510: call the current process recursively, that is, enter step 503.

Step 511: Exit. If the current procedure is one that was called recursively, fall back to the procedure that called the step.

Step 512: Write the mark "reference" at the page number of the current entry in the mulu.doc file.

Finally, the directory shown in Figure 3 is obtained.

Step 3: Construct the paper book body

The steps to construct the text of the paper book are shown in Figure 6:

Step 601: start. Create a file that records the text: content.doc. The page in content.doc is set to 5 areas, as shown in Figure 4, 401 is the header area, 402 is the text area, 403 is the right leaf node mark area, 405 is the text page number area, 406 is the left parent-child relationship mark area , where 404 is the mark area of the page number at the beginning of the knowledge point of the cited entry in the text area. Open the mulu.doc file, positioned at the beginning.

Step 602: Find the next entry, and find the record of the corresponding knowledge point in the file Knowledge.xls.

Step 603: Take out the mark at the page number of the current entry in the mulu.doc file.

Step 604: If the flag is "used", go to step 619, otherwise go to step 605.

Step 605: Obtain the start page number data of the corresponding knowledge point in the file Knowledge.xls.

Step 606: Add the symbol "*" in front of the page number data, and overwrite the mark data at the page number of the current entry in the mulu.doc file.

Step 607: Add the name of the current entry and its page number to the back of the file content.doc.

Step 608: If the current entry is a child node of the previous entry, go to step 609, otherwise go to step 610.

Step 609: Mark the symbol "┕" in the left parent-child relationship sign area of the file content.doc.

Step 610: mark the name of the knowledge point and the starting page number at the top of the current page of the content.doc file on the header.

Step 611: Go to the next page of the content.doc file.

Step 612: end.

Step 613 : Check if there are unprocessed entries in the mulu.doc file, if so, go to step 602 , otherwise go to step 612 .

Step 614: mark the symbol "■" in the right leaf node mark area of the file content.doc.

Step 615 : Check whether the current entry is a leaf node or a reference node, if yes, go to step 614 , otherwise, go to step 613 .

Step 616: Check whether the last page of the content.doc file is reached, if so, go to step 615, otherwise go to step 611.

Step 617: Arrive at the start page of the current knowledge point in the content.doc file.

Step 618: Overwrite the mark data at the page number of the current entry in the mulu.doc file with the page number data at the start position of the current knowledge point.

Step 619: Copy the data in the current knowledge point file to the end of the content.doc file, and record the page number at the start position of the knowledge point in the page number data item of Knowledge.xls.

Step 4: Build an index for knowledge points and other content retrieval

(4.1) Arrange all the knowledge point data in the Knowledge.xls file in ascending order of names, and copy the knowledge point names and their page numbers in the sorted result to the end of the content.doc file.

(4.2) Enter all the keywords and page numbers in the paper book into the Excel file for sorting, and then copy the sorting results to the end of the content.doc file.

Finally, get the table of contents of the paper book in mulu.doc, and get the text and index of the paper book in content.doc.

This specific implementation step is suitable for general paper book construction. For those of ordinary skill in the art, the methods involved in the present invention can be implemented in a general computer system, or can be manually programmed and implemented. For example, it can be implemented using Office and other software.

Those skilled in the art can easily understand that the above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention, All should be included within the protection scope of the present invention.

Claims (6)

1. the paper book construction method of a stratification explicit knowledge network, it is characterised in that described method comprises the steps:

(1) Knowledge Objects is built: the domain knowledge involved by paper book is resolved into a series of knowledge point being absent from Name Conflict；

(2) paper book catalogue is built: adopt Depth Priority Searching to organize formation knowledge main frame tree according to cause effect relation all of Knowledge Objects according to the theme of paper book, and show by bibliographic structure display packing；

(3) according to the directory tree that step (2) is obtained, from the beginning of catalogue, being described the knowledge point of correspondence entry in the content of the paper book text being written to order pagination, what also record each knowledge point in the body of the email starts the symbol of relation between the page number and some expression directory tree relevant entrys simultaneously；

(4) search method of knowledge point and other guide is built.

2. the method for claim 1, it is characterised in that described step (1) particularly as follows:

For each knowledge point, it is used that letter symbol some sections independent makes a distinction, namely each knowledge point is used that a knowledge point title and its corresponding assemble of symbol represent, described letter symbol includes graphical symbol and mathematical symbol, described knowledge point only comprises the description of current layer, a sub-knowledge point then row entry and do not show content.

3. method as claimed in claim 1 or 2, it is characterised in that described step (2) particularly as follows:

(2.1) the use mark of all of knowledge point is set to " use ", total for paper book title and total knowledge point are set to current entries, directory tree records current entries, and the use arranging the knowledge point of current entries and correspondence is masked as " use "；

(2.2) find and all of sub-knowledge point of current entries and the current knowledge point entry as current entries, and they are arranged according to certain order；

(2.3) all sub-entry to current entries, according to selected every the sub-entry that put in order set, is recorded on directory tree；

(2.4) if the use of this knowledge point is masked as " use ", then it is set using in this sub-entry and is masked as " quoting ", otherwise used mark to be set to " use "；

(2.5) if the use of this sub-entry knowledge point of there is sub-entry and its correspondence is masked as " use ", the knowledge point use then arranging correspondence is masked as " use ", and repeat (2.2)-(2.5), until the A to Z of point is all used, finally obtain a directory tree.

4. method as claimed in claim 1 or 2, it is characterised in that described step (3) particularly as follows:

(3.1) from directory tree, obtain an entry, it is set to current entries, find the knowledge point of correspondence.

(3.2) if the use mark of current entries is " use ", then the description of current knowledge point is added in the afterbody of the current text page of paper book, and this starting position, knowledge point is recorded at the page number place of current entries, if the use mark of current entries is " quoting ", then find the starting position page number of corresponding knowledge point, and this page number is write the page number place of current directory, and add " * " before the page number quoted and distinguished；

(3.3) in beginning one page of current knowledge point, if the previous entry of current entries is father node, then its beginning or left margin place mark this entry with on the filiation symbol " ┕ " of an entry, the relation represented between previous entry and current entries not marked is brotgher of node relation.

(3.4) in the text each page at current knowledge point place, if this knowledge point is in the forefront of current page, then the first page number of this knowledge point is marked at the header place of current page；

(3.5) in the last page of current knowledge point, if current entries is leaf node, then marking this knowledge point terminal " ■ " in its ending place or the right blank space, what do not mark is branch node, represents that part below is the child node of current knowledge point.

5. method as claimed in claim 1 or 2, it is characterised in that described step (4) particularly as follows:

(4.1) all of knowledge point title and the page number thereof are listed according to certain order order after paper book；

(4.2) other key words in all paper book and the page number thereof are listed according to certain order order after paper book.

6. method as claimed in claim 1 or 2, it is characterised in that described step (4) particularly as follows:

Other key words described refer to the concept term in paper book.