A knowledge graph fusion method, device and equipment based on two-dimensional image icons

Abstract

The application relates to a knowledge graph fusion method, device and equipment based on two-dimensional image icons, which comprises the following steps: for each first two-dimensional image icon in a first knowledge graph to be fused and each second two-dimensional image icon in a second knowledge graph, judging whether the two two-dimensional image icons belong to two-dimensional image icons corresponding to a same concept name according to the first two-dimensional image icon and the second two-dimensional image icon; if the two two-dimensional image icons belong to two-dimensional image icons corresponding to the same concept name, merging a first node and a second node corresponding to the two two-dimensional image icons into a target node, and merging the two two-dimensional image icons into a target two-dimensional image icon corresponding to the same concept name, so as to realize fusion of the two knowledge graphs. Through the method, the accuracy of knowledge graph fusion is improved, and the fusion efficiency is also improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of knowledge graph fusion, in particular, the present application relates to a knowledge graph fusion method, device and equipment based on two-dimensional image icon. BACKGROUND

[0002] The large-scale fusion technology of knowledge graph refers to the combination of two or more knowledge graphs into one, forming a higher level of knowledge expression. Knowledge graph fusion technology first needs to align concepts, that is, to determine whether the concepts in the concept graph refer to the same object. Taking tables as an example, it is necessary to first merge table headers and then perform entity fusion, that is, to copy table records in the corresponding positions.

[0003] Deficiencies of existing methods:

[0004] (1) It is impossible to determine whether the same concept is based on the concept word or word group:

[0005] General concepts are represented by words or word groups. However, due to different fields, different positions, and different scenarios, the same object is represented differently. For example, "buried steel pipeline" and "pipeline steel". The traditional method to distinguish between these two concepts is to give them attribute dimensions such as "material, weight, and purpose", etc. There are thousands of attributes. If these attributes are the same or more than 90%, it is considered to be the same concept. The graph method is to select the part of the graph where the two words are located. If the two parts of the graph are the same or more than 90%, it is considered to be the same concept. These methods of comparing concepts based on word attributes are all methods of operating within the same word set. According to the incompleteness theorem of Gödel, these methods of comparing words are incomplete, so theoretically, these methods cannot solve the problem of whether the concepts are the same.

[0006] (2) Concept graph relies on manual determination and is not suitable for graph fusion with more than 100 nodes:

[0007] Generally, whether two concept graphs are the same is determined by experts. This is generally only suitable for small-scale concepts, such as less than 100 concepts, in the same field. In the case of large-scale concepts, such as more than 100 concepts, there may be real-world problems such as cross-disciplinary, cross-industry, and cross-language. Experts also have knowledge limitations and cannot accurately determine whether concepts outside their cognitive range are the same. For example, "overprotection" refers to a small additional protection current in the field of overprotection, and refers to high voltage or current in the field of communication. The subject and phenomenon described by this word in the two fields are opposite, which makes it difficult for field experts to judge.

[0008] In summary, the two existing technologies cannot accurately determine whether two words belong to the same concept. SUMMARY

[0009] The technical problem solved by the present application is to provide a knowledge graph fusion method, device and equipment based on two-dimensional image icons, aiming to solve at least one of the above technical problems.

[0010] In a first aspect, the technical solution of the present application to solve the above technical problems is as follows: a knowledge graph fusion method based on two-dimensional image icons, the method comprising:

[0011] Obtaining a first knowledge graph and a second knowledge graph to be fused, for each knowledge graph in the first knowledge graph and the second knowledge graph, the knowledge graph includes a plurality of two-dimensional image icons, for each two-dimensional image icon, the two-dimensional image icon includes a concept name and an image icon, each two-dimensional image icon corresponds to a node in the knowledge graph, and in the knowledge graph, the membership relationship between nodes is represented by the connection between nodes.

[0012] For each first two-dimensional image icon in the first knowledge graph and each second two-dimensional image icon in the second knowledge graph, according to the first two-dimensional image icon and the second two-dimensional image icon, it is judged whether the first two-dimensional image icon and the second two-dimensional image icon belong to the same two-dimensional image icon corresponding to the concept name.

[0013] For each first two-dimensional image icon and each second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon belong to the same two-dimensional image icon corresponding to the concept name, then the first node corresponding to the first two-dimensional image icon and the second node corresponding to the second two-dimensional image icon are merged into a target node, and the first two-dimensional image icon and the second two-dimensional image icon are merged into a target two-dimensional image icon corresponding to the same concept name, so as to realize the fusion of the first knowledge graph and the second knowledge graph.

[0014] If the first two-dimensional image icon and the second two-dimensional image icon do not belong to the same two-dimensional image icon corresponding to the concept name, then the first knowledge graph and the second knowledge graph are not fused.

[0015] The beneficial effects of the present application are: in the process of graph fusion of the first knowledge graph and the second knowledge graph, the concept name is described by a two-dimensional image icon instead of the text description concept name in the prior art, which can obtain a way closer to human image thinking, expand the perception range of the concept, increase the accuracy of the consistency determination of different concept connotations, and on this basis, the fusion of the knowledge graphs can be realized based on whether the two-dimensional image icons of different knowledge graphs belong to the two-dimensional image icons corresponding to the same concept name, which not only improves the accuracy of the knowledge graph fusion, but also improves the fusion efficiency.

[0016] On the basis of the above technical solutions, the present application can also be improved as follows.

[0017] Further, for each first two-dimensional image icon in the first knowledge graph and each second two-dimensional image icon in the second knowledge graph, the above determining whether the first two-dimensional image icon and the second two-dimensional image icon belong to the two-dimensional image icons corresponding to the same concept name according to the first two-dimensional image icon and the second two-dimensional image icon comprises:

[0018] determining whether the first two-dimensional image icon and the second two-dimensional image icon are the same according to the first two-dimensional image icon and the second two-dimensional image icon;

[0019] if the first two-dimensional image icon and the second two-dimensional image icon are the same, determining that the first two-dimensional image icon and the second two-dimensional image icon belong to the two-dimensional image icons corresponding to the same concept name;

[0020] if the first two-dimensional image icon and the second two-dimensional image icon are not the same, determining that the first two-dimensional image icon and the second two-dimensional image icon do not belong to the two-dimensional image icons corresponding to the same concept name.

[0021] The beneficial effects of the above further scheme are that since the two-dimensional image icon is a picture, it can be determined whether the first two-dimensional image icon and the second two-dimensional image icon belong to the two-dimensional image icons corresponding to the same concept name by directly determining whether the first two-dimensional image icon and the second two-dimensional image icon are the same.

[0022] Further, if the first two-dimensional image icon and the second two-dimensional image icon are not the same, the method further comprises:

[0023] determining whether the first image icon corresponding to the first two-dimensional image icon and the second image icon corresponding to the second two-dimensional image icon are the same;

[0024] If the first image icon and the second image icon are the same, it is determined that the first concept name corresponding to the first image icon and the second concept name corresponding to the second image icon are synonyms, the first concept name is updated to the second two-dimensional image icon, and the second concept name is updated to the first two-dimensional image icon.

[0025] The beneficial effect of the above further scheme is that if the first two-dimensional image icon and the second two-dimensional image icon are not the same, but the first image icon and the second image icon are the same, it indicates that the first concept name corresponding to the first image icon and the second concept name corresponding to the second image icon are synonyms, so that the concept name and its corresponding synonyms can be represented by the same two-dimensional image icon.

[0026] Further, for each knowledge graph in the first knowledge graph and the second knowledge graph, the above knowledge graph further includes synonyms of each concept name, and the method further includes:

[0027] For each two-dimensional image icon in the knowledge graph, the synonyms corresponding to the concept name corresponding to the two-dimensional image icon are obtained according to the two-dimensional image icon.

[0028] The beneficial effect of the above further scheme is that since the knowledge graph further includes synonyms of each concept name, the synonyms corresponding to the concept name can also be obtained through the two-dimensional image icon.

[0029] Further, for each knowledge graph in the first knowledge graph and the second knowledge graph, the plurality of nodes in the knowledge graph include a starting point and an ending point, the first node is the starting point and / or the ending point, and the second node is the starting point and / or the ending point.

[0030] The beneficial effect of the above further scheme is that the knowledge graph includes a starting point and an ending point, and for each node in the first node and the second node, the node can not only be a starting point, but also an ending point, so that the target node after merging can not only represent a starting point or an ending point corresponding to the same concept name, but also represent a starting point and an ending point corresponding to the same concept name.

[0031] Further, for each first two-dimensional image icon and each second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name, the method further includes:

[0032] The first concept name corresponding to the first two-dimensional image icon is adjusted to the same concept name, and the second concept name corresponding to the second two-dimensional image icon is adjusted to the same concept name.

[0033] The beneficial effect of the further scheme is that when the first two-dimensional image icon and the second two-dimensional image icon are two-dimensional image icons corresponding to the same concept name, the first concept name corresponding to the first two-dimensional image icon in the first knowledge graph and the second concept name corresponding to the second two-dimensional image icon in the second knowledge graph can be adjusted according to the same concept name, so that the first concept name and the second concept name are both adjusted to the same concept name.

[0034] Further, for each two-dimensional image icon, the image icon corresponding to the two-dimensional image icon is displayed on the two-dimensional image icon.

[0035] The beneficial effect of the further scheme is that the image icon corresponding to the two-dimensional image icon is displayed on the two-dimensional image icon, so that people can easily construct the connotation of the concept and easily determine the similarities and differences between the two-dimensional image icons.

[0036] In a second aspect, the present application also provides a knowledge graph fusion device based on two-dimensional image icons, which comprises:

[0037] a knowledge graph acquisition module, configured to acquire a first knowledge graph and a second knowledge graph to be fused, for each of the first knowledge graph and the second knowledge graph, the knowledge graph comprising a plurality of two-dimensional image icons, for each two-dimensional image icon, the two-dimensional image icon comprising a concept name and an image icon, each two-dimensional image icon corresponding to a node in the knowledge graph, and in the knowledge graph, the membership relationship between nodes is represented by the connection between the nodes;

[0038] a first judgment module, configured to, for each first two-dimensional image icon in the first knowledge graph and each second two-dimensional image icon in the second knowledge graph, judge whether the first two-dimensional image icon and the second two-dimensional image icon are two-dimensional image icons corresponding to the same concept name according to the first two-dimensional image icon and the second two-dimensional image icon;

[0039] a fusion module, configured to, for each first two-dimensional image icon and each second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon are two-dimensional image icons corresponding to the same concept name, merge a first node corresponding to the first two-dimensional image icon and a second node corresponding to the second two-dimensional image icon into a target node, and merge the first two-dimensional image icon and the second two-dimensional image icon into a target two-dimensional image icon corresponding to the same concept name, so as to realize the fusion of the first knowledge graph and the second knowledge graph;

[0040] a processing module, configured to, when the first two-dimensional image icon and the second two-dimensional image icon are not two-dimensional image icons corresponding to the same concept name, not to perform fusion processing on the first knowledge graph and the second knowledge graph.

[0041] In a third aspect, the present application provides an electronic device to solve the above technical problems, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the knowledge graph fusion method based on two-dimensional iconic image when executing the computer program.

[0042] In a fourth aspect, the present application provides a computer readable storage medium to solve the above technical problems, which stores a computer program, and the computer program is executed by a processor to implement the knowledge graph fusion method based on two-dimensional iconic image.

[0043] Additional aspects and advantages of the application will be described in the following description and will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments of the present application will be briefly introduced.

[0045] Figure 1 A flowchart of a knowledge graph fusion method based on two-dimensional iconic image provided by an embodiment of the present application;

[0046] Figure 2 A schematic diagram of a knowledge graph establishment and knowledge graph fusion technology flow based on two-dimensional iconic image provided by an embodiment of the present application;

[0047] Figure 3 A schematic diagram of a knowledge graph fusion method based on two-dimensional iconic image provided by an embodiment of the present application;

[0048] Figure 4 A panoramic view of a concept graph represented by text provided by an embodiment of the present application;

[0049] Figure 5 A partial enlarged view of Figure 4 provided by an embodiment of the present application;

[0050] Figure 6 A schematic diagram of an iconic image provided by an embodiment of the present application;

[0051] Figure 7 A schematic diagram of a two-dimensional iconic image provided by an embodiment of the present application;

[0052] Figure 8 A display schematic diagram of information obtained by scanning a two-dimensional iconic image provided by an embodiment of the present application;

[0053] Figure 9 A two-dimensional image icon in a catalog provided by an embodiment of the present application;

[0054] Figure 10 A global map of a fused concept graph provided by an embodiment of the present application;

[0055] Figure 11 A partial enlarged view of Figure 10 provided by an embodiment of the present application;

[0056] Figure 12 A schematic diagram of a concept picture library provided by an embodiment of the present application;

[0057] Figure 13 A schematic diagram of a concept two-dimensional image icon library provided by an embodiment of the present application;

[0058] Figure 14 A result schematic diagram corresponding to an icon knowledge graph provided by an embodiment of the present application;

[0059] Figure 15 A structure schematic diagram of a table corresponding to a new graph provided by an embodiment of the present application;

[0060] Figure 16 A structure schematic diagram of a knowledge graph fusion device based on a two-dimensional image icon provided by an embodiment of the present application;

[0061] Figure 17 A structure schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0062] The principles and features of the present application are described below, and the examples are only used to explain the present application, and are not used to limit the scope of the present application.

[0063] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0064] The scheme provided by the embodiment of the present application can be applied to any application scenario that needs to be fused with a knowledge graph. The scheme provided by the embodiment of the present application can be executed by any electronic device, such as a terminal device of a user, including at least one of the following: a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, a smart television, and a smart vehicle device.

[0065] An embodiment of the present application provides a possible implementation manner, such as Figure 1 As shown in a flowchart of a knowledge graph fusion method based on a two-dimensional image icon, the scheme can be executed by any electronic device, for example, can be a terminal device, or executed by a terminal device and a server together. For the convenience of description, the method provided by the embodiment of the present application will be described below by taking the terminal device as an execution subject, as shown in the flowchart of Figure 1 The method can include the following steps:

[0066] Step S110, acquiring a first knowledge graph and a second knowledge graph to be fused. For each of the first knowledge graph and the second knowledge graph, the knowledge graph includes a plurality of two-dimensional image icons. For each two-dimensional image icon, the two-dimensional image icon includes a concept name and an image icon. Each two-dimensional image icon corresponds to a node in the knowledge graph. In the knowledge graph, the membership relationship between nodes is represented by a connection line between nodes.

[0067] Step S120, for each first two-dimensional image icon in the first knowledge graph and each second two-dimensional image icon in the second knowledge graph, determining whether the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name according to the first two-dimensional image icon and the second two-dimensional image icon.

[0068] Step S130, for each first two-dimensional image icon and each second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name, merging a first node corresponding to the first two-dimensional image icon and a second node corresponding to the second two-dimensional image icon into a target node, and merging the first two-dimensional image icon and the second two-dimensional image icon into a target two-dimensional image icon corresponding to the same concept name, so as to realize fusion of the first knowledge graph and the second knowledge graph.

[0069] Step S140, if the first two-dimensional image icon and the second two-dimensional image icon do not belong to two-dimensional image icons corresponding to the same concept name, not performing fusion processing on the first knowledge graph and the second knowledge graph.

[0070] By the method, in the process of graph fusion of the first knowledge graph and the second knowledge graph, the concept name is described by a two-dimensional image icon instead of a text description concept name in the prior art, so that a way closer to human image thinking is obtained, the perception range of the concept is expanded, the accuracy of the consistency determination of different concept connotations is increased, and on this basis, whether the two-dimensional image icons of different knowledge graphs are corresponding two-dimensional image icons of the same concept name is determined, the fusion of the knowledge graphs is realized, the accuracy of the knowledge graph fusion is improved, and the fusion efficiency is improved.

[0071] The scheme of the present application will be further described below in combination with the following specific embodiments, in which, referring to Figure 1 and Figure 2 the knowledge graph fusion method based on the two-dimensional image icon provided in the embodiment can include the following steps:

[0072] In step S110, a first knowledge graph and a second knowledge graph to be fused are acquired, for each of the first knowledge graph and the second knowledge graph, the knowledge graph includes a plurality of two-dimensional image icons, for each two-dimensional image icon, the two-dimensional image icon includes a concept name and an image icon, each two-dimensional image icon corresponds to a node in the knowledge graph, and in the knowledge graph, the membership relationship between nodes is represented by a connection line between nodes.

[0073] The two-dimensional image icon can be a two-dimensional code, i.e., a two-dimensional picture, a concept is a reflection of the essence of an objective thing by the human brain, is a knowledge unit formed by a unique combination of features, is a thinking unit reflecting common characteristics extracted from a group of things by using an abstract way, and different concepts can be distinguished by different concept names; an image is a more in-depth and objective and accurate meaning image formed by the brain through abstract thinking activities, is an objective corresponding entity of a vague concept; an image icon is to assign a corresponding object to an icon similar to a LOGO as a unique identifier of the concept; and a two-dimensional image icon of a concept is to transform the icon into a two-dimensional code, so as to facilitate the recognition of a computer or a smart device.

[0074] The connection between nodes can be directed, and the direction indicates the membership between the two connected nodes. For example, node A points to node B, which means that node A contains node B. In a knowledge graph, not all nodes are connected. Optionally, for each of the first knowledge graph and the second knowledge graph, the nodes in the knowledge graph can include a starting point and an ending point, and the connection between the nodes represents the membership between the starting point and the ending point. For a node, it can be both an ending point and a starting point. Generally, for a connected starting point and ending point, the ending point belongs to the starting point. The membership can be indicated by the direction of the connection, which is usually from the starting point to the ending point. As an example, see Figure 5 , the starting point is the natural gas pipeline ACI detection and protection report, the ending point is the detailed detection report of the third part of the natural gas pipeline ACI in Jiangxi Province, and the connection between the starting point and the ending point is from the starting point to the ending point. The membership between the starting point and the ending point is that the ending point belongs to the starting point.

[0075] Optionally, for each of the first knowledge graph and the second knowledge graph, the knowledge graph also includes synonyms of each concept name. The method further includes:

[0076] For each two-dimensional image icon in the knowledge graph, the synonyms corresponding to the concept name corresponding to the two-dimensional image icon are obtained according to the two-dimensional image icon.

[0077] The first knowledge graph and the second knowledge graph are both graphs that represent concepts by two-dimensional image icons. For a graph that represents concepts by text, it needs to be converted into a graph that represents concepts by two-dimensional image icons. The specific implementation process is as follows:

[0078] First, a two-dimensional image icon library of concepts is established. The representative concept pictures (also known as image icons) collected are abstracted and modified to make the pictures universal and general. Then, the picture is transformed into a two-dimensional picture code (i.e., a two-dimensional image icon), thereby simplifying the image recognition process. The two-dimensional image icon can also carry more text information.

[0079] Second, for a graph that represents concepts by text, a two-dimensional image icon is assigned to the graph. That is, each node in the graph is represented by a corresponding two-dimensional image icon. This way, a concept described by traditional words, word groups, or phrases has a symbolic two-dimensional icon, just like everyone has a two-dimensional face code. In this way, the graph that represents concepts by text is converted into a graph that represents concepts by two-dimensional image icons.

[0080] Optionally, the specific implementation process of converting the concept graph represented by text into the concept graph represented by two-dimensional image icons is as follows:

[0081] Referring to Figure 3 , the method corresponds to a technology composed of a data layer 1, a graph fusion layer 2, and an application layer 3, wherein the function of the data layer is to store various knowledge graphs and a conversion dictionary (which can also be referred to as an image icon dictionary), and to implement modification of the knowledge graphs and the conversion dictionary, wherein the conversion dictionary is a dictionary for converting the concept graph represented by text into the concept graph represented by two-dimensional image icons, which includes an image icon corresponding to each concept name in a concept name and a two-dimensional image icon, and then the corresponding two-dimensional image icon can be found in the conversion dictionary based on the concept name;

[0082] The graph fusion layer 2 is used to convert the general text knowledge graph (the concept graph represented by text) into the knowledge graph with two-dimensional image icons (the concept graph represented by two-dimensional image icons), which can be achieved by establishing a conversion from a concept picture library (each image icon) to a two-dimensional image icon library (the two-dimensional image icon corresponding to each image icon), and the conversion process can be automatically completed by a picture (image icon) to two-dimensional code python program; the application layer 3 is used to read in a new knowledge graph (as shown in Figure 3 ) and display the total knowledge graph after fusion by two-dimensional icons, which can also be referred to as a new knowledge graph or a fused knowledge graph.

[0083] Specifically, the data layer 1 includes a knowledge graph 1-1, a concept picture library 1-2, a concept two-dimensional image icon library 1-3, and a knowledge graph with icons 1-4. The knowledge graph 1-1 generally stores concept names represented by text in triples, and the triples include a starting point, an ending point, and a membership relationship between the starting point and the ending point, and the basic form is as shown in Figure 4 Figure 4 Each node can represent a concept name (a concept name represented by text), and one node corresponds to an ellipse in Figure 4 Figure 4 The panorama graph corresponding to the knowledge graph is mainly to display nodes with connection quantities from more to less from the center to the outside according to the hierarchy, and the number of connections is considered as an important indicator in the graph, Figure 5 Figure 4 is a local enlarged view of

[0084] The concept picture library 1-2 is used to correspond the concept name, the synonym of the concept name, and the image icon of the concept name, and the specific form of the concept picture library 1-2 can be seen from Table 1 as shown in Figure 12 , wherein, as an example, an image icon can be seen from Figure 6 .​​​Figure 6 The concept name is "Protection Measures for Continuous Interference of Natural Gas Pipeline", and the corresponding image icon is shown in the figure. The role of the concept two-dimensional image icon library 1-3 is to convert the image icon corresponding to the concept name into a two-dimensional code. For specific forms, please refer to Figure 13 Table 2 shown in the figure. For the two-dimensional image icon corresponding to the concept name, please refer to Figure 7 The two-dimensional image icon shown in the figure. The image icon can be displayed on the two-dimensional image icon, or the information of the image icon can be stored in the two-dimensional image icon. The two-dimensional image icon can also store the concept name (use the concept name as the icon name of the two-dimensional image icon) and the synonyms corresponding to the concept name. In actual application, after scanning the two-dimensional image icon by a smart device, the information stored in the two-dimensional image icon can be displayed, including the concept name and the synonyms of the concept name. For details, please refer to Figure 8 The information shown in the figure. The protection measures for continuous interference are the concept name, and the protection measures for continuous interference are the synonyms. If the image icon is stored in the two-dimensional image icon, the image icon can be displayed separately by a smart device. The display mode of the image icon is not limited and is within the protection scope of the present application. After obtaining multiple two-dimensional image icons, these two-dimensional image icons can be stored in a specified directory. For details, please refer to Figure 9 The two-dimensional image icon shown in the figure in the directory. Figure 9 As can be seen, the icon name corresponding to each two-dimensional image icon is the same as the concept name.

[0085] The role of the icon knowledge graph 1-4 is to give a two-dimensional image icon to all concepts on the basis of the original knowledge graph (a graph representing concepts in text). For example, Figure 14 Table 3 shown in the figure. The two fields of "all concepts" and "all concepts_two-dimensional icon" are added, that is, a two-dimensional image icon is given to each concept name. The picture and text information of the two-dimensional icon are used to identify the concept, which jumps out of the pure text field of the concept and is more in line with the image thinking mode of human beings.

[0086] The role of the above-mentioned graph fusion layer 2 is to realize the assignment of two-dimensional image icons to graphs and the adjustment of concepts according to image icons. For the two-dimensional image icons of the same concept, the concept name represented by the icon is used (for example, the concepts of "Nanchang-Gongqing Section 1st Stake AC Interference Mitigation Effect Detection Result", "Nanchang-Gongqing Section 15th Stake AC Interference Mitigation Effect Detection Result", and "Jiujiang-Gongqing Section 91st Stake AC Interference Mitigation Effect Detection Result" have the same icon of "On-site Implementation of Protection Scheme for AC Interference Seriousness", so the above-mentioned concepts are considered to be the same concept "On-site Implementation of Protection Scheme for AC Interference Seriousness"). Therefore, the concept graph fusion and reorganization based on two-dimensional image icons are realized.

[0087] The atlas fusion layer 2 includes four parts: assigning a two-dimensional code to text / picture 2-1, assigning an icon to a concept text 2-2, merging according to an icon node 2-3 (eliminating repeated entries according to a two-dimensional image icon), and constructing a merged atlas 2-4. The assigning a two-dimensional code to text / picture 2-1 refers to using the qrcode module of python (by saving the picture into a corresponding byte stream, and then transforming according to the two-dimensional code format) to transform the image icon corresponding to the concept into a two-dimensional code, adding the concept name and the data information of the unified concept (for example, data = field mitigation test at test pile No. 15 field mitigation test at test pile No. 15 field mitigation test at test pile No. 1, where “field mitigation test at test pile” represents a unified concept, and “field mitigation test at test pile No. 15” and “field mitigation test at test pile No. 1” are two entities or synonyms of it) in the two-dimensional code, and assigning an icon image icon (image icon) to the two-dimensional code image icon (image icon). Due to the introduction of the icon image (image icon), it is easy to construct the connotation of the concept and determine the similarities and differences between the concepts.

[0088] The assigning an icon to a concept text 2-2 has the effect of assigning an icon to all nodes in a concept atlas (an atlas representing concepts in text). According to the name of each node in the atlas, the corresponding two-dimensional image icon is found in the concept name and synonym field in Table 2 shown in Figure 13 , and the process of assigning a two-dimensional image icon to the concept is realized, so as to improve the concept described in text to the cognitive field of picture, increase the visual perception of the concept, and enrich the comprehensive understanding of the concept.

[0089] In the scheme of the present application, the merging according to an icon node 2-3 can also realize atlas fusion. The specific implementation process is as follows: the merging according to an icon node 2-3 refers to, in Table 3 shown in Figure 14 , for the nodes with two-dimensional image icons, the corresponding two-dimensional image icon concept name is used to replace the start and end names in the atlas, so as to realize the fusion of the concept nodes according to the two-dimensional image icons of the concepts; the effect of constructing a merged atlas 2-4 is to merge several atlases together to form a new atlas. The structure of the table corresponding to the new atlas can be seen from Table 4 shown in Figure 15 .

[0090] As an example, the result of merging two concept atlases “GBT 50698-2011 Buried Steel Pipeline AC Interference Protection Technical Standard.pdf” and “Jiangxi Natural Gas Pipeline AC Interference Detection and Protection Report V3.0.docx” two literature concept atlases, the global graph of the fused concept atlas is shown in Figure 10 , and the local enlarged view of the fused concept atlas is shown inFigure 11 As shown from the partial view, two knowledge graphs are fused through node 3, and according to Figure 15 The node 3 in Table 4 shown is obtained through adjustment of the two-dimensional iconic representation of the concept, that is, the node 3 represents the target node obtained after merging, and the node 3 can represent both the end point and the start point, and through the node 3, different nodes of the same concept are expressed, so that the fusion of knowledge graphs based on the two-dimensional iconic representation of the concept is realized.

[0091] The role of the above application layer 3 is to realize the loading of the graph (the graph representing the concept in words) and the display of the fused graph, including two modules of reading the knowledge graph 3-1 and displaying the merged graph 3-2. The reading knowledge graph 3-1 is used to realize the loading of the knowledge graph, and the knowledge graph file of the triple is read into the software platform; the display of the merged graph 3-2 completes the visualization of the merged knowledge graph, and the visualization result is as shown in Figure 11 and Figure 12 .

[0092] In step S120, for each first two-dimensional iconic representation in the first knowledge graph and each second two-dimensional iconic representation in the second knowledge graph, it is judged whether the first two-dimensional iconic representation and the second two-dimensional iconic representation belong to the two-dimensional iconic representations corresponding to the same concept name according to the first two-dimensional iconic representation and the second two-dimensional iconic representation.

[0093] Wherein, whether the first two-dimensional iconic representation and the second two-dimensional iconic representation belong to the two-dimensional iconic representations corresponding to the same concept name refers to whether the two two-dimensional iconic representations from two different knowledge graphs express the same concept name.

[0094] Optionally, step S120 specifically includes:

[0095] According to the first two-dimensional iconic representation and the second two-dimensional iconic representation, it is judged whether the first two-dimensional iconic representation and the second two-dimensional iconic representation are the same; that is, it is judged whether the two icons are the same from the image comparison level (appearance).

[0096] If the first two-dimensional iconic representation and the second two-dimensional iconic representation are the same, it is judged that the first two-dimensional iconic representation and the second two-dimensional iconic representation belong to the two-dimensional iconic representations corresponding to the same concept name.

[0097] If the first two-dimensional iconic representation and the second two-dimensional iconic representation are not the same, it is judged that the first two-dimensional iconic representation and the second two-dimensional iconic representation do not belong to the two-dimensional iconic representations corresponding to the same concept name.

[0098] Since the two-dimensional image icon also includes an image icon and a concept name, it is not accurate to determine whether two two-dimensional image icons are the same based on the two-dimensional image icon itself alone, so if the first two-dimensional image icon and the second two-dimensional image icon are not the same, the method further includes:

[0099] determining whether the first image icon corresponding to the first two-dimensional image icon and the second image icon corresponding to the second two-dimensional image icon are the same;

[0100] If the first image icon and the second image icon are the same, it is determined that the first concept name corresponding to the first image icon and the second concept name corresponding to the second image icon are synonyms and belong to the same concept, so the first concept name is updated to the second two-dimensional image icon, and the second concept name is updated to the first two-dimensional image icon, so that the first concept name and its synonym the second concept name have the same two-dimensional image icon.

[0101] Step S130, for each of the first two-dimensional image icon and each of the second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon are two-dimensional image icons corresponding to the same concept name, the first node corresponding to the first two-dimensional image icon and the second node corresponding to the second two-dimensional image icon are merged into a target node, and the first two-dimensional image icon and the second two-dimensional image icon are merged into a target two-dimensional image icon corresponding to the same concept name, to realize the fusion of the first knowledge graph and the second knowledge graph;

[0102] Step S140, if the first two-dimensional image icon and the second two-dimensional image icon are not two-dimensional image icons corresponding to the same concept name, the first knowledge graph and the second knowledge graph are not fused.

[0103] In the process of fusing the first knowledge graph and the second knowledge graph, for the first two-dimensional image icon and the second two-dimensional image icon which are not two-dimensional image icons corresponding to the same concept name, the first two-dimensional image icon and the second two-dimensional image icon which do not belong to the same concept name can not be changed in the fused graph, and become the two-dimensional image icon corresponding to the node in the fused graph.

[0104] For the fused graph, the nodes corresponding to the same concept and the corresponding two-dimensional image icons are merged, that is, the essence of fusion is to represent the nodes representing the same concept by the same node and represent the two-dimensional image icons representing the same concept by the same two-dimensional image icon, so as to reduce some redundant icons and nodes, and make the fused graph more clearly express each node and the relationship between the nodes. The specific implementation process of the above fusion processing can be that all nodes in the two knowledge graphs are first merged to obtain all nodes in the fused graph, then the corresponding two-dimensional image icon of each node in all nodes is determined, and for the two-dimensional image icon of the node that needs to be adjusted, the two-dimensional image icon of the same concept name is adjusted from the two-dimensional image icon before merging, that is, the node name of each node in all nodes is adjusted, and the node name can be represented by a concept name.

[0105] In addition, if the fusion process of the first knowledge graph and the second knowledge graph is based on the starting points and the ending points in the graph, the starting points representing the same concept in the two knowledge graphs are merged, the ending points representing the same concept in the two knowledge graphs are merged, and the starting points and the ending points representing the same concept in the two knowledge graphs are merged. Further, in the fusion process of the first knowledge graph and the second knowledge graph, all starting points and ending points in the two knowledge graphs can be first fused together, then the starting points and / or ending points that change in the first knowledge graph are adjusted, and the starting points and / or ending points that change in the second knowledge graph are adjusted by analogy. The graph structure does not change, but from the fused graph, the originally separated starting points or ending points are merged into one node, thereby realizing the fusion of the knowledge graph based on the two-dimensional icon. For details, see Figure 15 The table 4 shown in the table 4 is boxed, which is the starting point and the ending point before and after adjustment.

[0106] Optionally, for the fused graph, visual display can be performed. It can also be applied to different application scenarios, for example, based on the fused knowledge graph, a search application is constructed to improve the credibility of search results; an intelligent question and answer is constructed to solve the attribute migration problem between cross-node long-range joint nodes and improve the experience of question and answer. It should be noted that the knowledge graph in the scheme of the present application is not limited to reflecting the knowledge of which field, and the scheme can be exemplified by the pipe network field for specific description.

[0107] Optionally, for each of the first two-dimensional image icon and the second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon are two-dimensional image icons corresponding to the same concept name, the method further includes:

[0108] Adjust the first concept name corresponding to the first two-dimensional image icon as the same concept name, and adjust the second concept name corresponding to the second two-dimensional image icon as the same concept name.

[0109] The same concept name can have the following cases: the same concept name can be the same as the first concept name, the same concept name can be the same as the second concept name, the same concept name can be a synonym of the first concept name, and the same concept name can be a synonym of the second concept name, wherein the first concept name and the second concept name can be the same or synonyms of each other.

[0110] Optionally, for any one of the first knowledge graph and the second knowledge graph, a node in the knowledge graph can be a starting point or a terminal point. Therefore, for a first node in the first knowledge graph, the first node can be a starting point or a terminal point, or both. For a second node in the second knowledge graph, the second node can be a starting point or a terminal point, or both. Similarly, for each node in the fused graph, the node can be a starting point or a terminal point, or both.

[0111] As an example, see Figure 5 For the concept name corresponding to node 1 of the 3rd part of the detailed detection report of the natural gas pipeline in Jiangxi Province, the node 1 is a starting point relative to the concept name corresponding to node 2 of the detailed detection conclusion and suggestion of 3.4, and the node 1 is a terminal point relative to the concept name corresponding to node 3 of the detection and protection of the natural gas pipeline, then the node 1 is both a terminal point and a starting point.

[0112] In order to better illustrate and understand the principles of the method provided by the present application, the scheme of the present application will be described below in combination with an optional specific embodiment. It should be noted that the specific implementation of each step in the specific embodiment should not be understood as a limitation of the scheme of the present application. Other implementation manners that can be thought of by those skilled in the art on the basis of the principles of the scheme provided by the present application should also be regarded as within the scope of protection of the present application.

[0113] In the present example, see Figure 2 how the first knowledge graph and the second knowledge graph are converted, and the graph fusion process of the first knowledge graph and the second knowledge graph, which specifically includes the following steps:

[0114] First, how the first knowledge graph and the second knowledge graph are converted, since the conversion processes of the two knowledge graphs are different Figure 2The concept-image icon processing) is the same, the following is a specific description of the first knowledge graph as an example:

[0115] Step 1, read the concept_image picture file

[0116] The concept_image picture file is a form composed of three fields of concept name, synonym and image picture shown in Table 2 shown in Table 2, wherein the image picture (i.e. image icon) records the location and name of the picture, which is the original understanding of the concept. Figure 13

[0117] Step 2, prepare qrcode (software) data / icon

[0118] The text information (data) in the two-dimensional code is composed of concept name + synonym, and through scanning the two-dimensional code, the various names of the concept under the same concept can be known; icon is the image picture;

[0119] Step 3, construct a two-dimensional code

[0120] According to the requirements of the two-dimensional code (according to the field requirements of the qrcode software module), the text information and picture information (image picture) are integrated together.

[0121] Step 3-1, determine the basic parameters of the two-dimensional code

[0122] Select version number 5, error correction setting high, two-dimensional code control box (box_size) 8 pixels, box border (border) 1 pixel to construct an empty two-dimensional code qr; The above several parameters are the parameters required for converting the above text information and picture information into a two-dimensional code based on the Python qrcode two-dimensional code module, and the specific use of the Python qrcode can be referred to in the prior art, which will not be repeated here.

[0123] Step 3-2, load data

[0124] The prepared data (such as data = test pile site mitigation test @ 15 test pile site mitigation test # 1 test pile site mitigation test) is loaded onto the empty two-dimensional code through the data adding function (qr.add_data());

[0125] Step 3-3, icon format preprocessing

[0126] ​Set the color and position parameters of the image in the QR code using the image generation function (img=qr.make_image()) and the image color conversion function (img=img.convert("RGBA")). Then open the image file, put the image into the QR code image, remove the excess parts, and pad the insufficient parts with zeros.

[0127] Steps 3-4: Paste the icon onto the QR code.

[0128] By using the icon pasting function `img.paste(icon,(w,h),icon)` on the QR code image, you can paste the processed icon into the image area of ​​the QR code to obtain a QR code with an image.

[0129] Steps 3-5: Save the conceptual two-dimensional image icon.

[0130] Save the converted icon (two-dimensional image icon) in Figure 13 The table in Table 2 shows a series of two-dimensional iconographic representations of the concepts. Table 2 provides a common graphical basis for the integration of different knowledge graphs.

[0131] After constructing the first and second knowledge graphs based on the above steps, graph fusion is performed on the first and second knowledge graphs. Figure 2 The following is an introduction to the knowledge graph fusion process shown:

[0132] Step 4, Read the knowledge graph file

[0133] The original knowledge graph (a graph that represents concepts in words) was created by Figure 14 The start-end-relationship triplet shown in Table 3 is used to read this file into memory for processing.

[0134] Step 5: Merge the start and end points of the map.

[0135] The total concept node is obtained by merging all nodes at the starting point and ending point (two points on any line segment in the graph) in the first and second knowledge graphs.

[0136] Step 6: Assign icons to all concepts

[0137] The task of assigning icons to all concepts is accomplished by searching a dictionary of conceptual icon representations (which can be a pre-built graph of concepts using two-dimensional icon representations based on steps 1 to 4 above). Specifically, for each concept name corresponding to each node in the merged set of nodes, the corresponding two-dimensional icon is searched from the icon dictionary. The result is as follows: Figure 14 Table 3 shows the results after assigning icons, where the two additional fields, "All Concepts" and "All Concepts_2D Icons", represent the final result.

[0138] Step 7, adjust the starting point according to the concept icon

[0139] This is the first ring of the completion of the merger, which is achieved by changing the node name corresponding to the starting point (which can be the same as the concept name) to the corresponding concept name corresponding to the two-dimensional iconic image in Table 3 (for example, the starting point "5 Communication Interference Protection Measures" is replaced by "Communication Interference Protection Scheme Design Specification"), so that different concepts are changed to the same concept name representation, and the graph structure does not change, but from the graph, the originally separated starting points are merged into one node, thereby realizing the fusion of the knowledge graph according to the two-dimensional icon.

[0140] Step 8, adjust the end point according to the concept icon

[0141] This is the second ring of the completion of the merger, which is achieved by changing the node name of the end point (which can be the same as the concept name) to the corresponding concept name corresponding to the two-dimensional iconic image in Table 3 (for example, the end point "5.1 General Provisions" is replaced by "General Provisions for Communication Interference Protection Measures"), so that different concepts are changed to the same concept name representation, and the graph structure does not change, but from the graph, the originally separated end points are merged into one node, thereby realizing the fusion of the knowledge graph according to the two-dimensional icon.

[0142] Step 9, construct the adjusted graph file

[0143] The adjusted knowledge graph file is shown in Table 4 as shown in Figure 15 .

[0144] Step 10, knowledge graph visualization

[0145] The (adjusted_starting_point, adjusted_ending_point, relationship) in Table 4 forms the fused new knowledge graph triple, and the graph is displayed through the visualization tool.

[0146] The scheme of the present application fuses large-scale knowledge graphs through two-dimensional iconic images of concepts, breaks out of the traditional way of describing concepts in text, and expands the perception range of concepts by giving concepts two-dimensional iconic images, thereby increasing the accuracy of determining the consistency of different concept connotations. In the fusion of a graph of 1151 concepts and a graph of 1260 concepts, traditional expert determination requires 10 days, while the iconic image method can achieve fusion in only 1 day, greatly reducing the cognitive difficulty of large-scale graph fusion.

[0147] Based on the Figure 1The embodiments of the present application also provide a knowledge graph fusion device 20 based on two-dimensional image icons according to the same principle as the method shown in the background art, as shown in Figure 16 The knowledge graph fusion device 20 based on two-dimensional image icons can include a knowledge graph acquisition module 210, a first judgment module 220, a fusion module 230 and a processing module 240, as shown in the background art, wherein:

[0148] The knowledge graph acquisition module 210 is configured to acquire a first knowledge graph and a second knowledge graph to be fused. For each of the first knowledge graph and the second knowledge graph, the knowledge graph includes a plurality of two-dimensional image icons. For each of the two-dimensional image icons, the two-dimensional image icon includes a concept name and an image icon. Each of the two-dimensional image icons corresponds to a node in the knowledge graph. In the knowledge graph, the membership relationship between nodes is represented by the connection between nodes.

[0149] The first judgment module 220 is configured to, for each of a first two-dimensional image icon in the first knowledge graph and a second two-dimensional image icon in the second knowledge graph, judge whether the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name according to the first two-dimensional image icon and the second two-dimensional image icon.

[0150] The fusion module 230 is configured to, for each of the first two-dimensional image icon and each of the second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name, merge a first node corresponding to the first two-dimensional image icon and a second node corresponding to the second two-dimensional image icon into a target node, and merge the first two-dimensional image icon and the second two-dimensional image icon into a target two-dimensional image icon corresponding to the same concept name, so as to realize the fusion of the first knowledge graph and the second knowledge graph.

[0151] The processing module 240 is configured to, if the first two-dimensional image icon and the second two-dimensional image icon do not belong to two-dimensional image icons corresponding to the same concept name, not perform the fusion processing on the first knowledge graph and the second knowledge graph.

[0152] Optionally, for each of the first two-dimensional image icon in the first knowledge graph and each of the second two-dimensional image icon in the second knowledge graph, the first judgment module 220 is specifically configured to:

[0153] According to the first two-dimensional image icon and the second two-dimensional image icon, it is judged whether the first two-dimensional image icon and the second two-dimensional image icon are the same; if the first two-dimensional image icon and the second two-dimensional image icon are the same, it is judged that the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name; if the first two-dimensional image icon and the second two-dimensional image icon are not the same, it is judged that the first two-dimensional image icon and the second two-dimensional image icon do not belong to two-dimensional image icons corresponding to the same concept name.

[0154] Optionally, if the first two-dimensional image icon and the second two-dimensional image icon are not the same, the device further comprises:

[0155] The second judging module is configured to judge whether the first image icon corresponding to the first two-dimensional image icon and the second image icon corresponding to the second two-dimensional image icon are the same; if the first image icon and the second image icon are the same, it is judged that the first concept name corresponding to the first image icon and the second concept name corresponding to the second image icon are synonyms, the first concept name is updated to the second two-dimensional image icon, and the second concept name is updated to the first two-dimensional image icon.

[0156] Optionally, for each of the first knowledge graph and the second knowledge graph, the knowledge graph further comprises synonyms of each of the concept names, and the device further comprises:

[0157] The synonym obtaining module is configured to, for each of the two-dimensional image icons in the knowledge graph, obtain synonyms of the concept name corresponding to the two-dimensional image icon according to the two-dimensional image icon.

[0158] Optionally, for each of the first knowledge graph and the second knowledge graph, the plurality of nodes in the knowledge graph comprise a starting point and an ending point, the first node is the starting point and / or the ending point, and the second node is the starting point and / or the ending point.

[0159] Optionally, for each of the first two-dimensional image icon and each of the second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name, the device further comprises:

[0160] The adjusting module is configured to adjust the first concept name corresponding to the first two-dimensional image icon to the same concept name and adjust the second concept name corresponding to the second two-dimensional image icon to the same concept name.

[0161] Optionally, for each of the two-dimensional image icons, the image icon corresponding to the two-dimensional image icon is displayed on the two-dimensional image icon.

[0162] The two-dimensional image icon-based knowledge graph fusion device provided in the embodiments of the present application can execute the two-dimensional image icon-based knowledge graph fusion method provided in the embodiments of the present application, and the implementation principles are similar. The actions performed by each module and unit in the two-dimensional image icon-based knowledge graph fusion device in the embodiments of the present application are corresponding to the steps in the two-dimensional image icon-based knowledge graph fusion method in the embodiments of the present application. The detailed function description of each module of the two-dimensional image icon-based knowledge graph fusion device can be found in the description of the corresponding two-dimensional image icon-based knowledge graph fusion method shown in the foregoing, which will not be described here.

[0163] The two-dimensional image icon-based knowledge graph fusion device can be a computer program (including program code) running in a computer device, for example, the two-dimensional image icon-based knowledge graph fusion device is an application software. The device can be used to execute the corresponding steps in the method provided in the embodiments of the present application.

[0164] In some embodiments, the two-dimensional image icon-based knowledge graph fusion device provided in the embodiments of the present application can be implemented in a combination of software and hardware. For example, the two-dimensional image icon-based knowledge graph fusion device provided in the embodiments of the present application can be a hardware decoding processor in the form of a processor programmed to execute the two-dimensional image icon-based knowledge graph fusion method provided in the embodiments of the present application. For example, the hardware decoding processor in the form of a processor can use one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs), or other electronic elements.

[0165] In some other embodiments, the two-dimensional image icon-based knowledge graph fusion device provided in the embodiments of the present application can be implemented in a software manner, Figure 16The two-dimensional image icon-based knowledge graph fusion device stored in the memory can be software in the form of programs and plug-ins, and includes a series of modules, including an access request acquisition module 210, a target webpage package determination module 220, and a two-dimensional image icon-based knowledge graph fusion module 230, for implementing the two-dimensional image icon-based knowledge graph fusion method provided in the embodiments of the present application.

[0166] The modules described in the embodiments of the present application can be implemented in the form of software or hardware. In some cases, the names of the modules do not limit the modules themselves.

[0167] Based on the same principles as the method shown in the embodiments of the present application, the embodiments of the present application also provide an electronic device, which can include but is not limited to a processor and a memory; the memory is configured to store a computer program; and the processor is configured to execute the method shown in any of the embodiments of the present application by invoking the computer program.

[0168] In an optional embodiment, an electronic device is provided, as shown in Figure 17 , as shown in Figure 17 The electronic device 4000 shown in the embodiments of the present application includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, through a bus 4002. Optionally, the electronic device 4000 can also include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not limit the embodiments of the present application.

[0169] The processor 4001 can be a CPU (Central Processing Unit, central processing unit), a general-purpose processor, a DSP (Digital Signal Processor, digital signal processor), an ASIC (Application Specific Integrated Circuit, application specific integrated circuit), an FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure. The processor 4001 can also be a combination of computing functions, such as one or more microprocessor combinations, DSP and microprocessor combinations, etc.

[0170] The bus 4002 can include a path that transmits information between the above-described components. The bus 4002 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, or the like. The bus 4002 can be divided into an address bus, a data bus, a control bus, or the like. For convenience of representation, Figure 17 Only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0171] The memory 4003 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0172] The memory 4003 is used to store application program code (computer program) for implementing the scheme of the present application, and is controlled by the processor 4001 to execute. The processor 4001 is used to execute the application program code stored in the memory 4003 to realize the content shown in the foregoing method embodiments.

[0173] The electronic device can also be a terminal device, Figure 17 The electronic device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0174] The embodiments of the present application provide a computer readable storage medium, and the computer readable storage medium stores a computer program. When the computer program runs on a computer, the computer can execute the corresponding content in the foregoing method embodiments.

[0175] According to another aspect of the present application, there is also provided a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the methods provided in the various embodiments.

[0176] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0177] It should be understood that the flowchart and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of various embodiments of the present application. In this regard, each block in the flowchart and block diagrams can represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or combinations of hardware and software.

[0178] The above computer readable storage medium stores one or more programs, which, when executed by the electronic device, cause the electronic device to perform the methods described in the above embodiments.

[0179] The above description is only the preferred embodiment of the present application and the explanation of the applied technical principles. It should be understood by those skilled in the art that the disclosed range of the present application is not limited to the technical scheme formed by the specific combination of the above technical features, and should also cover other technical schemes formed by the combination of the above technical features or equivalent features without departing from the disclosed concept. For example, the technical scheme formed by the mutual replacement of the above features and the technical features with similar functions disclosed in the present application (but not limited to) without departing from the disclosed concept.

Claims

1. A knowledge graph fusion method based on two-dimensional image icons, characterized in that, The method comprises the following steps: acquiring a first knowledge graph and a second knowledge graph to be fused, for each of the first knowledge graph and the second knowledge graph, the knowledge graph comprises a plurality of two-dimensional image icons, for each of the two-dimensional image icons, the two-dimensional image icon comprises a concept name and an image icon, each of the two-dimensional image icons corresponds to a node in the knowledge graph, and a membership relationship between nodes is represented by a connection line between the nodes in the knowledge graph; for each of a first two-dimensional image icon in the first knowledge graph and a second two-dimensional image icon in the second knowledge graph, determining whether the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to a same concept name according to the first two-dimensional image icon and the second two-dimensional image icon; for each of the first two-dimensional image icon and the second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to a same concept name, merging a first node corresponding to the first two-dimensional image icon and a second node corresponding to the second two-dimensional image icon into a target node, and merging the first two-dimensional image icon and the second two-dimensional image icon into a target two-dimensional image icon corresponding to the same concept name, so as to realize fusion of the first knowledge graph and the second knowledge graph; if the first two-dimensional image icon and the second two-dimensional image icon do not belong to two-dimensional image icons corresponding to a same concept name, not performing fusion processing on the first knowledge graph and the second knowledge graph; for each of the first two-dimensional image icon in the first knowledge graph and the second two-dimensional image icon in the second knowledge graph, the determining whether the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to a same concept name according to the first two-dimensional image icon and the second two-dimensional image icon comprises: determining whether the first two-dimensional image icon and the second two-dimensional image icon are the same according to the first two-dimensional image icon and the second two-dimensional image icon; if the first two-dimensional image icon and the second two-dimensional image icon are the same, determining that the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to a same concept name; if the first two-dimensional image icon and the second two-dimensional image icon are not the same, determining that the first two-dimensional image icon and the second two-dimensional image icon do not belong to two-dimensional image icons corresponding to a same concept name; if the first two-dimensional image icon and the second two-dimensional image icon are not the same, the method further comprises: determining whether a first image icon corresponding to the first two-dimensional image icon and a second image icon corresponding to the second two-dimensional image icon are the same; If the first image icon and the second image icon are the same, it is determined that the first concept name corresponding to the first image icon and the second concept name corresponding to the second image icon are synonyms, the first concept name is updated to the second two-dimensional image icon, and the second concept name is updated to the first two-dimensional image icon.

2. The method of claim 1, wherein, For each of the knowledge graphs in the first knowledge graph and the second knowledge graph, the knowledge graph further includes synonyms of each of the concept names, and the method further includes: For each of the two-dimensional image icons in the knowledge graph, the synonyms corresponding to the concept name corresponding to the two-dimensional image icon are obtained according to the two-dimensional image icon.

3. The method of claim 1, wherein, For each of the knowledge graphs in the first knowledge graph and the second knowledge graph, the nodes in the knowledge graph include a starting point and an ending point, the first node is the starting point and / or the ending point, and the second node is the starting point and / or the ending point.

4. The method of claim 1, wherein, For each of the first two-dimensional image icon and the second two-dimensional image icon, if the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name, the method further includes: The first concept name corresponding to the first two-dimensional image icon is adjusted to the same concept name, and the second concept name corresponding to the second two-dimensional image icon is adjusted to the same concept name.

5. The method of claim 1, wherein, For each of the two-dimensional image icons, the image icon corresponding to the two-dimensional image icon is displayed on the two-dimensional image icon. 6.A knowledge graph fusion device based on two-dimensional image icons, characterized in that, The device comprises: A knowledge graph acquisition module is configured to acquire a first knowledge graph and a second knowledge graph to be fused, for each of the knowledge graphs in the first knowledge graph and the second knowledge graph, the knowledge graph includes a plurality of two-dimensional image icons, for each of the two-dimensional image icons, the two-dimensional image icon includes a concept name and an image icon, each of the two-dimensional image icons corresponds to a node in the knowledge graph, and in the knowledge graph, the membership relationship between nodes is represented by a connection line between nodes. A first judgment module is configured to, for each of the first two-dimensional image icons in the first knowledge graph and each of the second two-dimensional image icons in the second knowledge graph, determine whether the first two-dimensional image icon and the second two-dimensional image icon belong to two-dimensional image icons corresponding to the same concept name according to the first two-dimensional image icon and the second two-dimensional image icon. a fusion module configured to, for each of the first two-dimensional iconic icons and each of the second two-dimensional iconic icons, if the first two-dimensional iconic icon and the second two-dimensional iconic icon are two-dimensional iconic icons corresponding to a same concept name, merge a first node corresponding to the first two-dimensional iconic icon and a second node corresponding to the second two-dimensional iconic icon into a target node, and merge the first two-dimensional iconic icon and the second two-dimensional iconic icon into a target two-dimensional iconic icon corresponding to the same concept name, so as to realize fusion of the first knowledge graph and the second knowledge graph; a processing module configured to, if the first two-dimensional iconic icon and the second two-dimensional iconic icon are not two-dimensional iconic icons corresponding to a same concept name, not perform fusion processing on the first knowledge graph and the second knowledge graph.

7. An electronic device, comprising: A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method in any one of claims 1-5.

Images