CN112381661B

CN112381661B - Insurance product similarity determination method and apparatus

Info

Publication number: CN112381661B
Application number: CN202011354965.2A
Authority: CN
Inventors: 唐远洋; 欧阳凯; 陈健; 邹阳; 李思雯
Original assignee: Shenzhen Huize Times Technology Co ltd
Current assignee: Shenzhen Huize Times Technology Co ltd
Priority date: 2020-11-27
Filing date: 2020-11-27
Publication date: 2024-01-30
Anticipated expiration: 2040-11-27
Also published as: CN112381661A

Abstract

The invention discloses a method and a device for determining similarity of insurance products, which can determine that an insurance information node which is once associated with a first insurance product node and is not once associated with a second insurance product node in an insurance knowledge graph is a first difference information node, and determine that an insurance information node which is once associated with the second insurance product node and is not once associated with the first insurance product node in the insurance knowledge graph is a second difference information node, so that the difference between the first insurance product node and the second insurance product node can be embodied: the connection weights corresponding to the edges of the first insurance product nodes and the first difference information nodes and the connection weights corresponding to the edges of the second insurance product nodes and the second difference information nodes enable the similarity of the insurance products corresponding to the first insurance product nodes and the insurance products corresponding to the second insurance product nodes to be determined rapidly and accurately.

Description

Insurance product similarity determination method and apparatus

Technical Field

The invention relates to the technical field of computers, in particular to a method and a device for determining similarity of insurance products.

Background

With the improvement of the public's insurance consciousness, insurance products gradually become important guarantee products of the public for risks possibly born in the future.

As a special product, the insurance product has complexity far exceeding that of common daily products. An insurance product may include a number of insurance information such as insurance conditions, insurance liability terms, disease information associated with the insurance product, insurance general knowledge information, underwriting information, security information, and insurance company information. When people select insurance products, the people usually search for several similar insurance products for comparison and select the insurance products suitable for themselves from the comparison. However, in the face of massive insurance products and insurance information, even a person having a certain insurance knowledge has difficulty in accurately judging the similarity between the insurance products in a short time, thereby picking out similar insurance products.

Therefore, how to quickly and accurately determine the similarity between insurance products becomes a technical problem that needs to be solved by technicians.

Disclosure of Invention

In view of the above problems, the present invention provides a method and an apparatus for determining similarity of insurance products, which overcome or at least partially solve the above problems, and the technical solution is as follows:

A method for determining similarity of insurance products, comprising:

determining a first difference information node corresponding to a first insurance product node in an insurance knowledge graph, wherein the first difference information node is an insurance information node which is associated with the first insurance product node and is not associated with a second insurance product node in the insurance knowledge graph, the insurance knowledge graph is provided with an insurance product node, an insurance information node and edges connected with the nodes, the edges correspond to preset connection weights, each insurance product node corresponds to one insurance product, and each insurance information node corresponds to one insurance information;

determining a second difference information node corresponding to the second insurance product node in the insurance knowledge graph, wherein the second difference information node is an insurance information node which is associated with the second insurance product node once and is not associated with the first insurance product node once in the insurance knowledge graph;

and determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node at least according to the connection weight corresponding to the edge, connected with the first difference information node, of the first insurance product node and the connection weight corresponding to the edge, connected with the second difference information node, of the second insurance product node.

Optionally, the determining, at least according to the connection weight corresponding to the edge where the first insurance product node is connected to each first difference information node and the connection weight corresponding to the edge where the second insurance product node is connected to each second difference information node, the similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node includes:

determining a first negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the connection weight corresponding to the edge, connected with each first difference information node, of the first insurance product node;

determining a second negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the connection weight corresponding to the edge, connected with each second difference information node, of the second insurance product node;

and determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node at least according to the first negative similarity and the second negative similarity.

Optionally, the method further comprises:

determining a common information node on a connecting path between the first insurance product node and the second insurance product node in the insurance knowledge graph, wherein the first insurance product node and the second insurance product node are connected through at least one common information node, and the common information node is an insurance information node connected between the first insurance product node and the second insurance product node;

according to the connection path between the first insurance product node and the second insurance product node: determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, wherein the first connection weight corresponds to the edge connected with the common information node, the second connection weight corresponds to the edge connected with the common information node, and the third connection weight corresponds to the edge connected between different common information nodes;

the determining, at least according to the first negative similarity and the second negative similarity, the similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node includes:

And determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the first negative similarity, the second negative similarity and the positive similarity.

Optionally, the determining, according to the connection weight corresponding to the edge where the first insurance product node is connected to each first difference information node, the first negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node includes:

according to the formula:

determining a first negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, wherein value _A For the first negative similarity, a is the number of edges connected with the first security product node and each first difference information node, i is the braiding of the edges connected with the first security product node and the first difference information nodeA number; value _i The connection weight corresponding to the edge numbered i.

Optionally, the determining, according to the connection weight corresponding to the edge where the second insurance product node is connected to each second difference information node, the second negative similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node includes:

According to the formula:

determining a second negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, wherein value _B B is the number of edges, connected with the second difference information nodes, of the second insurance product node, j is the number of edges, connected with the second difference information nodes, of the second insurance product node; value _j The connection weight corresponding to the edge with the number j.

Optionally, the determining a common information node on a connection path between the first insurance product node and the second insurance product node in the insurance knowledge graph includes:

determining at least one connection path between the first insurance product node and the second insurance product node in the insurance knowledge graph;

determining the insurance information node on the at least one connection path as a common information node;

on the connection path between the first insurance product node and the second insurance product node: the determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node includes:

For any connection path on the at least one connection path: determining a first connection weight corresponding to the edge connected with the common information node of the first insurance product node, a second connection weight corresponding to the edge connected with the common information node of the second insurance product node and a third connection weight corresponding to the edge connected with different common information nodes on the connection path; determining sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path according to the first connection weight, the second connection weight and the third connection weight on the connection path;

and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the sub forward similarity on each connection path.

Optionally, the determining, according to the first connection weight, the second connection weight and the third connection weight on the connection path, the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path includes:

According to the formula:

determining the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path, wherein g is the number of the connection path, value _g For the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path with the number g, M is the number of edges connected with each common information node on the connection path, and M is the number of edges connected with each common information node on the connection pathEdge numbering, value _m And the connection weight corresponding to the edge numbered m on the connection path is obtained.

Optionally, the determining, according to the sub-forward similarity on each connection path, the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node includes:

according to the formula:

determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, wherein value _AB And G is the number of connection paths between the first insurance product node and the second insurance product node in the insurance knowledge graph for the forward similarity.

An insurance product similarity determining device, comprising: a first difference information node determining unit, a second difference information node determining unit, and an insurance product similarity determining unit,

the first difference information node determining unit is configured to determine a first difference information node corresponding to a first insurance product node in an insurance knowledge graph, where the first difference information node is an insurance information node that is associated with the first insurance product node and is not associated with a second insurance product node in the insurance knowledge graph, the insurance knowledge graph has an insurance product node, an insurance information node, and edges connected to the nodes, the edges correspond to a preset connection weight, each insurance product node corresponds to one insurance product, and each insurance information node corresponds to one insurance information;

the second difference information node determining unit is configured to determine a second difference information node corresponding to the second insurance product node in the insurance knowledge graph, where the second difference information node is an insurance information node that is once associated with the second insurance product node and is not once associated with the first insurance product node in the insurance knowledge graph;

The insurance product similarity determining unit is configured to determine similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node according to at least a connection weight corresponding to an edge, where the first insurance product node is connected to each first difference information node, and a connection weight corresponding to an edge, where the second insurance product node is connected to each second difference information node.

Optionally, the insurance product similarity determining unit includes: a first negative similarity determination subunit, a second negative similarity determination subunit, and an insurance product similarity determination subunit,

the first negative similarity determining subunit is configured to determine, according to a connection weight corresponding to an edge where the first insurance product node is connected to each first difference information node, a first negative similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node;

the second negative similarity determining subunit is configured to determine, according to a connection weight corresponding to an edge where the second insurance product node is connected to each second difference information node, a second negative similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node;

And the insurance product similarity determining subunit is configured to determine, at least according to the first negative similarity and the second negative similarity, a similarity between an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

By means of the technical scheme, the insurance product similarity determining method and device provided by the invention can determine that the insurance information node which is associated with the first insurance product node and is not associated with the second insurance product node in the insurance knowledge graph is the first difference information node, and determine that the insurance information node which is associated with the second insurance product node and is not associated with the first insurance product node in the insurance knowledge graph is the second difference information node, so that the difference between the first insurance product node and the second insurance product node can be embodied: the connection weights corresponding to the edges of the first insurance product nodes and the first difference information nodes and the connection weights corresponding to the edges of the second insurance product nodes and the second difference information nodes enable the similarity of the insurance products corresponding to the first insurance product nodes and the insurance products corresponding to the second insurance product nodes to be determined rapidly and accurately.

The foregoing description is only an overview of the present invention, and is intended to be implemented in accordance with the teachings of the present invention in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present invention more readily apparent.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a method for determining similarity of insurance products according to an embodiment of the present invention;

fig. 2 shows a schematic diagram of connection between a security product node and a security information node in a security knowledge graph according to an embodiment of the present invention;

FIG. 3 is a schematic flow chart of another method for determining similarity of insurance products according to an embodiment of the present invention;

FIG. 4 is a schematic flow chart of another method for determining similarity of insurance products according to an embodiment of the present invention;

Fig. 5 is a schematic structural diagram of an insurance product similarity determining device according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of another insurance product similarity determining device according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

As shown in fig. 1, a method for determining similarity of insurance products according to an embodiment of the present invention may include:

s100, determining a first difference information node corresponding to a first insurance product node in an insurance knowledge graph, wherein the first difference information node is an insurance information node which is associated with the first insurance product node at one time and is not associated with a second insurance product node at one time in the insurance knowledge graph.

The Knowledge Graph (knowledgegraph) can describe Knowledge resources and carriers thereof by using a visualization technology, and excavate, analyze, construct, draw and display Knowledge and interrelationships among the Knowledge resources and carriers. The insurance knowledge graph in the embodiment of the invention is provided with insurance product nodes, insurance information nodes and edges connected with the nodes, wherein the edges correspond to preset connection weights.

Optionally, the embodiment of the invention can pre-construct an insurance knowledge graph.

The embodiment of the invention can define the ontology concept of each insurance product in the insurance knowledge graph, wherein each insurance product node corresponds to one insurance product. For example, the node "Dart 3" represents the insurance product "Dart 3".

The embodiment of the invention can define the ontology concept of each insurance information in the insurance knowledge graph, wherein each insurance information node corresponds to one insurance information. Optionally, the body of the insurance product may include: the system comprises a guarantee responsibility entity, an insurance condition entity, a disease entity, an insurance general knowledge entity, a nuclear insurance entity, a security entity and an insurance company entity. For example: a "serious disease responsibility" node corresponding to the guarantee responsibility entity, a "thyroid nodule" node corresponding to the disease entity, and the like.

The embodiment of the invention can define the relation between the ontology and the ontology in the security knowledge graph. Specifically, the embodiment of the invention can input the edges representing the relationship among the nodes corresponding to each body when the insurance knowledge graph is constructed. For example: the entity of the insurance product and the entity of the insurance information have a 'responsibility-containing' relationship, and the insurance product node corresponding to the insurance product and the insurance information node corresponding to the insurance information in the insurance knowledge graph can be connected by the 'responsibility-containing' edge.

According to the embodiment of the invention, the corresponding connection weights can be respectively set for each side in the insurance knowledge graph. Optionally, the embodiment of the present invention may be applied to any security product node: setting the sum of the connection weights of the edges of the insurance information node once associated with the insurance product node to a fixed value, for example: the fixed value is 100. Optionally, the embodiment of the invention can set the connection weight of the edge connected between different insurance information nodes according to the experience of a technician. It will be appreciated that the magnitude of the connection weight of an edge may reflect the degree of tightness between two nodes to which the edge is connected.

Wherein a degree of association refers to a direct connection between two nodes. The first differential information node is an insurance information node directly connected with the first insurance product node product and not directly connected with the second insurance product node. For ease of understanding, the first degree association and the first difference information node are described herein in connection with fig. 2: as shown in fig. 2, the first insurance product node is directly connected with the insurance information node q, the insurance information node e and the insurance information node r, i.e. the first insurance product node is once connected with the insurance information node q, the insurance information node e and the insurance information node r. The second insurance product node is directly connected with the insurance information node w, the insurance information node e and the insurance information node y, namely the first insurance product node is once connected with the insurance information node w, the insurance information node e and the insurance information node y. And the insurance information node q and the insurance information node r are once associated with the first insurance product node and are not once associated with the second insurance product node, and the insurance information node q and the insurance information node r are first difference information nodes. Although insurance information node e is once associated with the first insurance product node, insurance information node e is also once associated with the second insurance product node, and thus insurance information node e is not taken as the first difference information node.

It can be appreciated that, in a general case, the embodiment of the present invention may determine at least one first difference information node in the insurance knowledge graph.

S200, determining a second difference information node corresponding to the second insurance product node in the insurance knowledge graph, wherein the second difference information node is an insurance information node which is associated with the second insurance product node in the insurance knowledge graph at one time and is not associated with the first insurance product node at one time.

The second differential information node is an insurance information node directly connected with the second insurance product node and not directly connected with the first insurance product node. The second difference information node is also explained based on fig. 2: as shown in fig. 2, if the insurance information node w and the insurance information node y are once associated with the second insurance product node and not once associated with the first insurance product node, the insurance information node w and the insurance information node y are second difference information nodes. Although the insurance information node e is once associated with the second insurance product node, the insurance information node e is also once associated with the first insurance product node, and thus the insurance information node e is not regarded as the second difference information node.

S300, determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node at least according to the connection weight corresponding to the edge, where the first insurance product node is connected with each first difference information node, and the connection weight corresponding to the edge, where the second insurance product node is connected with each second difference information node.

The connection weight corresponding to the edge of the first insurance product node connected with each first difference information node can be used as the data attribute of the first insurance product node for representing the self-insurance data characteristic in the insurance knowledge graph. The connection weight corresponding to the edge of the second insurance product node connected with each second difference information node can be used as the data attribute of the second insurance product node for representing the self-insurance data characteristic in the insurance knowledge graph. The embodiment of the invention can determine the degree of difference between the two insurance product nodes through the comparison of the data attributes of the two insurance product nodes, thereby determining the similarity between the two insurance products corresponding to the two insurance product nodes respectively.

Optionally, in the embodiment of the present invention, the similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node may be determined according to an average value of sums of connection weights corresponding to edges connected to the first insurance product node and the first difference information nodes, and an average value of sums of connection weights corresponding to edges connected to the second insurance product node and the second difference information nodes. Optionally, in the embodiment of the present invention, the sum of the two average values may be used as the similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node. Optionally, in the embodiment of the present invention, an average value of a sum of the two average values may be used as a similarity between an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

Optionally, based on the method shown in fig. 1, as shown in fig. 3, another method for determining similarity of insurance products provided in an embodiment of the present invention, step S300 may include:

s310, determining a first negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the connection weight corresponding to the edge, where the first insurance product node and each first difference information node are connected.

Optionally, the embodiment of the present invention may determine, according to a negative value of a sum of connection weights corresponding to edges where the first insurance product node is connected to each first difference information node, a first negative similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

Specifically, step S310 may include:

according to the formula:

determining a first negative similarity of an insurance product corresponding to a first insurance product node and an insurance product corresponding to a second insurance product node, wherein value _A For the first negative similarity, a is the number of edges connected with each first difference information node by the first insurance product node, i is the number of edges connected with the first difference information node by the first insurance product node; value _i The connection weight corresponding to the edge numbered i.

S320, determining a second negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the connection weight corresponding to the edge, where the second insurance product node is connected with each second difference information node.

Optionally, the embodiment of the present invention may determine, according to a negative value of a sum of connection weights corresponding to edges where the second insurance product node is connected to the second difference information nodes, a second negative similarity between an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

Specifically, step S320 may include:

according to the formula:

determining a second negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, wherein the value _B B is the edge of the second insurance product node connected with each second difference information nodeJ is the number of the edge connected with the second insurance product node and the second difference information node; value _j The connection weight corresponding to the edge with the number j.

S330, determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node at least according to the first negative similarity and the second negative similarity.

Optionally, according to the embodiment of the present invention, the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node may be determined according to the sum of the first negative similarity and the second negative similarity.

Optionally, in the embodiment of the present invention, the similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node may be determined according to the product of the first negative similarity and the second negative similarity.

According to the embodiment of the invention, the degree of difference between the first insurance product node and the second insurance product node can be judged through the first negative similarity and the second negative similarity, and the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node can be determined according to the degree of difference. According to the embodiment of the invention, the similarity between the insurance products corresponding to the insurance product nodes can be more scientifically and accurately determined through the difference degrees between the different insurance product nodes.

The insurance product similarity determining method provided by the invention can determine that the insurance information node which is once associated with the first insurance product node and is not once associated with the second insurance product node in the insurance knowledge graph is the first difference information node, and determine that the insurance information node which is once associated with the second insurance product node and is not once associated with the first insurance product node in the insurance knowledge graph is the second difference information node, so that the difference between the first insurance product node and the second insurance product node can be embodied: the connection weights corresponding to the edges of the first insurance product nodes and the first difference information nodes and the connection weights corresponding to the edges of the second insurance product nodes and the second difference information nodes enable the similarity of the insurance products corresponding to the first insurance product nodes and the insurance products corresponding to the second insurance product nodes to be determined rapidly and accurately.

Optionally, based on the method shown in fig. 3, as shown in fig. 4, another method for determining similarity of insurance products provided in an embodiment of the present invention may further include:

s400, determining a common information node on a connecting path between the first insurance product node and the second insurance product node in the insurance knowledge graph, wherein the first insurance product node and the second insurance product node are connected through at least one common information node, and the common information node is an insurance information node connected between the first insurance product node and the second insurance product node.

The connecting path is composed of a common information node between two insurance product nodes and an edge connecting the two insurance product nodes and the common information node between the two insurance product nodes. For a better understanding of connection paths and commonalities, the description herein is in connection with fig. 2: as shown in fig. 2, there are 3 connection paths between the first insurance product node and the second insurance product node, which are respectively "first insurance product node-insurance information node q-insurance information node w-second insurance product node", "first insurance product node-insurance information node e-second insurance product node", and "first insurance product node-insurance information node r-insurance information node t-insurance information node y-second insurance product node", and the insurance information node q, insurance information node w, insurance information node e, insurance information node r, insurance information node t, and insurance information node y in the three connection paths are all common information nodes between the first insurance product node and the second insurance product node. It will be appreciated that in the usual case, there is at least one connection path between two insurance product nodes.

Specifically, step S400 may include: at least one connection path between the first insurance product node and the second insurance product node in the insurance knowledge graph is determined. And determining the insurance information node on the at least one connection path as a common information node.

From the above description of the difference information node and the common information node, it is understood that in some cases, the same insurance information node may be the difference information node and the common information node at the same time.

S500, according to the connection path between the first insurance product node and the second insurance product node: and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the first connection weight corresponding to the edge connected with the common information node, the second connection weight corresponding to the edge connected with the common information node and the third connection weight corresponding to the edge connected between different common information nodes.

Because the first insurance product node and the second insurance product node are connected through the connection path, the connection weight corresponding to the edge on the connection path represents the commonality degree between the first insurance product node and the second insurance product node to a certain extent. The embodiment of the invention can determine the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node by using the connection weight of the edge on the connection path.

Optionally, the embodiment of the present invention may be based on a connection path between the first insurance product node and the second insurance product node: and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node by the average value of the sum of the first connection weight, the second connection weight and the third connection weight.

Optionally, the embodiment of the present invention may be based on a connection path between the first insurance product node and the second insurance product node: and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node.

Optionally, the embodiment of the present invention may be based on a connection path between the first insurance product node and the second insurance product node: and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node by the minimum value among the first connection weight, the second connection weight and the third connection weight.

The embodiment of the invention passes through the connection path between the first insurance product node and the second insurance product node: the first connection weight, the second connection weight and the third connection weight can judge the degree of commonality between the first insurance product node and the second insurance product node, and the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node can be scientifically and accurately determined according to the degree of commonality.

Optionally, step S500 may include:

for any connection path on the at least one connection path: and determining a first connection weight corresponding to the edge connected with the common information node of the first insurance product node, a second connection weight corresponding to the edge connected with the common information node of the second insurance product node and a third connection weight corresponding to the edge connected with different common information nodes on the connection path.

And determining the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path according to the first connection weight, the second connection weight and the third connection weight on the connection path.

Alternatively, embodiments of the present invention may be according to the formula:

determining a sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path, wherein the sub-forward similarity is determined by the sub-forward similarityWherein g is the number of the connection path, value _g For the sub-forward similarity of the first insurance product node and the second insurance product node on the connection path with the number g, M is the number of the edges connected with the common information nodes on the connection path, M is the number of the edges connected with the common information nodes on the connection path, and value _m And the connection weight corresponding to the edge numbered m on the connection path is obtained.

For example: the connection path of the first insurance product node and the second insurance product node with the number of 1 is 'first insurance product node-common information node 1-common information node 2-second insurance product node', the connection weight corresponding to the edge connected with the common information node 1 on the connection path with the number of 1 is 3, the connection weight corresponding to the edge connected with the common information node 1 and the common information node 2 is 1, and the connection weight corresponding to the edge connected with the common information node 2 and the second insurance product node is 3, and the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path with the number of 1 is 1.75.

Optionally, in the embodiment of the present invention, the sub-forward similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node may be determined on the connection path according to the sum of the first connection weight, the second connection weight and the third connection weight on the connection path.

Optionally, in the embodiment of the present invention, the sub-forward similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node may be determined on the connection path according to an average value of the sum of the first connection weight, the second connection weight and the third connection weight on the connection path.

Optionally, in the embodiment of the present invention, the sub-forward similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node may be determined on the connection path according to the maximum value among the first connection weight, the second connection weight and the third connection weight on the connection path.

Optionally, in the embodiment of the present invention, the sub-forward similarity between the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node may be determined on the connection path according to a minimum value among the first connection weight, the second connection weight and the third connection weight on the connection path.

And determining the forward similarity of the first insurance product node and the second insurance product node according to the sub forward similarity on each connection path.

Optionally, the embodiment of the present invention may determine the forward similarity between the first insurance product node and the second insurance product node according to the sum of the sub forward similarities on the connection paths.

Specifically, the embodiment of the present invention may be according to the formula:

determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, wherein the value _AB And G is the number of connection paths between the first insurance product node and the second insurance product node in the insurance knowledge graph for the forward similarity.

For example: the first insurance product node and the second insurance product node have a connection path with the number of 2 and a connection path with the number of 3, the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path with the number of 2 is 9, and the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path with the number of 3 is 7, so that the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node is 16.

Optionally, the embodiment of the present invention may determine the forward similarity between the first insurance product node and the second insurance product node according to an average value of sums of sub forward similarities on each connection path.

Optionally, the embodiment of the present invention may determine the forward similarity between the first insurance product node and the second insurance product node according to the maximum value among the sub forward similarities on each connection path.

Optionally, the embodiment of the present invention may determine the forward similarity between the first insurance product node and the second insurance product node according to a minimum value among the sub forward similarities on each connection path.

As shown in fig. 4, in another method for determining similarity of insurance products provided in an embodiment of the present invention, step S330 may include:

s331, determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the first negative similarity, the second negative similarity and the positive similarity.

As can be seen from the above description, the first negative similarity and the second negative similarity may represent the degree of difference between the first insurance product node and the second insurance product node, and the positive similarity may represent the degree of commonality between the first insurance product node and the second insurance product node.

Optionally, in the embodiment of the present invention, the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node may be determined according to the sum of the first negative similarity, the second negative similarity and the positive similarity.

similarity _AB ＝value _AB +value _A +value _B

and determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node. Wherein the similarity is _AB For the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, value _A For the first negative similarity, value _B Value for the second negative similarity _AB Is the forward similarity.

For example: when the first negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node is-17, the second negative similarity is-11, and the positive similarity is not 93.6, the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node is 65.6.

Because the first negative-direction similarity and the second negative-direction similarity are negative values, and the positive-direction similarity is a positive value, the degree of commonality and the degree of difference between the first insurance product node and the second insurance product node can be compared by summing the first negative-direction similarity, the second negative-direction similarity and the positive-direction similarity, and the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node can be further determined more scientifically and accurately.

According to the method for determining the similarity of the insurance products, when a user is interested in a certain insurance product, the similarity of the insurance product and other insurance products is determined through the method for determining the similarity of the insurance products, and further the other insurance products can be ranked according to the similarity from large to small, and the other insurance products ranked in front of a preset sequence position are recommended to the user, so that convenience is brought to the user to select the insurance products suitable for the user after the user compares the similar insurance products.

It will be appreciated that the various steps described in embodiments of the invention may be performed in a different order and/or performed in parallel. Fig. 4 shows one alternative execution sequence of step S400 and step S500. For example: in addition to the illustration of fig. 4, step S400 and step S500 may be performed simultaneously with step S100, or performed before step S100, etc. The execution sequence of step S400 and step S500 is not further limited in the embodiment of the present invention.

Corresponding to the above method embodiment, the embodiment of the present invention further provides an insurance product similarity determining device, where the structure of the insurance product similarity determining device is shown in fig. 5, and the insurance product similarity determining device may include: the first difference information node determining unit 100, the second difference information node determining unit 200, and the insurance product similarity determining unit 300.

The first difference information node determining unit 100 is configured to determine a first difference information node corresponding to a first insurance product node in an insurance knowledge graph, where the first difference information node is an insurance information node that is associated with the first insurance product node and is not associated with a second insurance product node in the insurance knowledge graph.

The embodiment of the invention can define the ontology concept of each insurance product in the insurance knowledge graph, wherein each insurance product node corresponds to one insurance product.

The embodiment of the invention can define the ontology concept of each insurance information in the insurance knowledge graph, wherein each insurance information node corresponds to one insurance information. Optionally, the body of the insurance product may include: the system comprises a guarantee responsibility entity, an insurance condition entity, a disease entity, an insurance general knowledge entity, a nuclear insurance entity, a security entity and an insurance company entity.

The embodiment of the invention can define the relation between the ontology and the ontology in the security knowledge graph. Specifically, the embodiment of the invention can input the edges representing the relationship among the nodes corresponding to each body when the insurance knowledge graph is constructed.

Wherein a degree of association refers to a direct connection between two nodes. The first differential information node is an insurance information node directly connected with the first insurance product node product and not directly connected with the second insurance product node.

It is understood that the first difference information node determining unit 100 may determine at least one first difference information node in the insurance knowledge graph in a general case.

The second difference information node determining unit 200 is configured to determine a second difference information node corresponding to the second insurance product node in the insurance knowledge graph, where the second difference information node is an insurance information node that is associated with the second insurance product node and is not associated with the first insurance product node in the insurance knowledge graph.

The second differential information node is an insurance information node directly connected with the second insurance product node and not directly connected with the first insurance product node.

The insurance product similarity determining unit 300 is configured to determine the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node at least according to the connection weight corresponding to the edge where the first insurance product node is connected to each first difference information node and the connection weight corresponding to the edge where the second insurance product node is connected to each second difference information node.

Optionally, the insurance product similarity determining unit 300 includes: the first negative similarity determination subunit, the second negative similarity determination subunit, and the insurance product similarity determination subunit.

And the first negative similarity determining subunit is used for determining the first negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the connection weight corresponding to the edge, connected with each first difference information node, of the first insurance product node.

Optionally, the first negative similarity determining subunit may determine, according to a negative value of a sum of connection weights corresponding to edges of the first insurance product node and the first difference information nodes, a first negative similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

Optionally, the first negative similarity determining subunit may be specifically configured to:

And the second negative similarity determining subunit is used for determining the second negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the connection weight corresponding to the edge, connected with each second difference information node, of the second insurance product node.

Optionally, the second negative similarity determining subunit may determine, according to a negative value of a sum of connection weights corresponding to edges where the second insurance product node is connected to the second difference information nodes, a second negative similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

Optionally, the second negative-going similarity determining subunit may be specifically configured to:

determining a second negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, wherein the value _B B is the number of edges connected with each second difference information node of the second insurance product node, j is the number of edges connected with the second difference information node of the second insurance product node; value _j The connection weight corresponding to the edge with the number j.

And the insurance product similarity determining subunit is used for determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node at least according to the first negative similarity and the second negative similarity.

Optionally, the insurance product similarity determining subunit may determine, according to a sum of the first negative similarity and the second negative similarity, a similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

Optionally, the insurance product similarity determining subunit may determine, according to a product of the first negative similarity and the second negative similarity, a similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

The insurance product similarity determining device provided by the invention can determine that the insurance information node which is once associated with the first insurance product node and is not once associated with the second insurance product node in the insurance knowledge graph is the first difference information node, and determine that the insurance information node which is once associated with the second insurance product node and is not once associated with the first insurance product node in the insurance knowledge graph is the second difference information node, so that the difference between the first insurance product node and the second insurance product node can be embodied: the connection weights corresponding to the edges of the first insurance product nodes and the first difference information nodes and the connection weights corresponding to the edges of the second insurance product nodes and the second difference information nodes enable the similarity of the insurance products corresponding to the first insurance product nodes and the insurance products corresponding to the second insurance product nodes to be determined rapidly and accurately.

Optionally, based on the apparatus shown in fig. 5, as shown in fig. 6, another apparatus for determining similarity of insurance products provided in an embodiment of the present invention may further include: a common information node determining unit 400 and a forward similarity determining unit 500.

The common information node determining unit 400 is configured to determine a common information node on a connection path between the first insurance product node and the second insurance product node in the insurance knowledge graph, where the first insurance product node and the second insurance product node are connected by at least one common information node, and the common information node is an insurance information node connected between the first insurance product node and the second insurance product node.

The connecting path is composed of a common information node between two insurance product nodes and an edge connecting the two insurance product nodes and the common information node between the two insurance product nodes.

Optionally, the commonality information node determining unit 400 may be specifically configured to determine at least one connection path between the first insurance product node and the second insurance product node in the insurance knowledge graph. And determining the insurance information node on the at least one connection path as a common information node.

A forward similarity determining unit 500, configured to, according to a connection path between the first insurance product node and the second insurance product node: and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the first connection weight corresponding to the edge connected with the common information node, the second connection weight corresponding to the edge connected with the common information node and the third connection weight corresponding to the edge connected between different common information nodes.

Alternatively, the forward similarity determining unit 500 may determine that the first insurance product node is connected to the second insurance product node according to the connection path between the first insurance product node and the second insurance product node: and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node by the average value of the sum of the first connection weight, the second connection weight and the third connection weight.

Alternatively, the forward similarity determining unit 500 may determine that the first insurance product node is connected to the second insurance product node according to the connection path between the first insurance product node and the second insurance product node: and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node.

Alternatively, the forward similarity determining unit 500 may determine that the first insurance product node is connected to the second insurance product node according to the connection path between the first insurance product node and the second insurance product node: and determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node by the minimum value among the first connection weight, the second connection weight and the third connection weight.

Alternatively, the forward similarity determination unit 500 may include a sub-forward similarity determination subunit and a forward similarity determination subunit.

A sub-forward similarity determining subunit, configured to, for any connection path on the at least one connection path: and determining a first connection weight corresponding to the edge connected with the common information node of the first insurance product node, a second connection weight corresponding to the edge connected with the common information node of the second insurance product node and a third connection weight corresponding to the edge connected with different common information nodes on the connection path. And determining the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path according to the first connection weight, the second connection weight and the third connection weight on the connection path.

Alternatively, the sub-forward similarity determination subunit may be specifically configured to:

determining sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path, wherein g is the number of the connection path, value _g For the sub-forward similarity of the first insurance product node and the second insurance product node on the connection path with the number g, M is the number of the edges connected with the common information nodes on the connection path, M is the number of the edges connected with the common information nodes on the connection path, and value _m On the connection pathThe connection weight corresponding to the edge with the number m.

Optionally, the sub-forward similarity determining subunit may determine, on the connection path, a sub-forward similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node according to a sum of the first connection weight, the second connection weight, and the third connection weight on the connection path.

Optionally, the sub-forward similarity determining subunit may determine, on the connection path, a sub-forward similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node according to an average value of a sum of the first connection weight, the second connection weight, and the third connection weight on the connection path.

Optionally, the sub-forward similarity determining subunit may determine, on the connection path, a sub-forward similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node according to a maximum value among the first connection weight, the second connection weight, and the third connection weight on the connection path.

Optionally, the sub-forward similarity determining subunit may determine, on the connection path, a sub-forward similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node according to a minimum value among the first connection weight, the second connection weight, and the third connection weight on the connection path.

And the forward similarity determination subunit is used for determining the forward similarity of the first insurance product node and the second insurance product node according to the sub forward similarity on each connection path.

Optionally, the forward similarity determining subunit may determine the forward similarity of the first insurance product node and the second insurance product node according to a sum of the sub forward similarities on the connection paths.

Specifically, the forward similarity determination subunit may be specifically configured to:

Optionally, the forward similarity determining subunit may determine the forward similarity of the first insurance product node and the second insurance product node according to an average value of sums of the sub forward similarities on the connection paths.

Optionally, the forward similarity determining subunit may determine the forward similarity of the first insurance product node and the second insurance product node according to a maximum value among the sub forward similarities on each connection path.

Optionally, the forward similarity determining subunit may determine the forward similarity of the first insurance product node and the second insurance product node according to a minimum value among the sub forward similarities on each connection path.

The insurance product similarity determining unit 300 may be specifically configured to determine, according to the first negative similarity, the second negative similarity, and the positive similarity, a similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node.

Optionally, the insurance product similarity determining unit 300 may determine the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the sum of the first negative similarity, the second negative similarity and the positive similarity.

similarity _AB ＝value _AB +value _A +value _B

The insurance product similarity determining device comprises a processor and a memory, wherein the first difference information node determining unit 100, the second difference information node determining unit 200, the insurance product similarity determining unit 300 and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.

The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be provided with one or more than one, and the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node can be rapidly and accurately determined by adjusting the kernel parameters.

The embodiment of the invention provides a storage medium, on which a program is stored, which when executed by a processor, implements the insurance product similarity determination method.

The embodiment of the invention provides a processor which is used for running a program, wherein the program runs to execute the insurance product similarity determining method.

The embodiment of the invention provides electronic equipment, which comprises at least one processor, and at least one memory and a bus which are connected with the processor; the processor and the memory complete communication with each other through a bus; the processor is used for calling program instructions in the memory to execute the insurance product similarity determination method. The electronic device herein may be a server, a PC, a PAD, a mobile phone, etc.

The present application also provides a computer program product adapted to perform a program initialized with the above-mentioned insurance product similarity determining method steps when executed on an electronic device.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus, electronic devices (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, the electronic device includes one or more processors (CPUs), memory, and a bus. The electronic device may also include input/output interfaces, network interfaces, and the like.

The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip. Memory is an example of a computer-readable medium.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims

1. A method for determining similarity of insurance products, comprising:

2. The method of claim 1, wherein determining the similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node based at least on the connection weights corresponding to the edges of the first insurance product node and the first difference information node and the connection weights corresponding to the edges of the second insurance product node and the second difference information node, comprises:

3. The method as recited in claim 2, further comprising:

determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the first connection weight corresponding to the edge connected with the common information node, the second connection weight corresponding to the edge connected with the common information node and the third connection weight corresponding to the edge connected with different common information nodes on the connection path between the first insurance product node and the second insurance product node;

4. The method of claim 2, wherein determining the first negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the connection weight corresponding to the edge of the first insurance product node connected to each of the first difference information nodes comprises:

according to the formula:

determining a first negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node, wherein value _a For the first negative similarity, a is the number of edges connected with the first difference information nodes by the first insurance product node, i is the number of edges connected with the first difference information nodes by the first insurance product node; value _i The connection weight corresponding to the edge numbered i.

5. The method of claim 2, wherein determining the second negative similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the connection weight corresponding to the edge of the second insurance product node connected to each of the second difference information nodes comprises:

According to the formula:

6. A method according to claim 3, wherein said determining a commonality information node on a connection path between the first insurance product node and the second insurance product node in the insurance knowledge graph comprises:

the determining, according to a first connection weight corresponding to an edge of the first insurance product node connected with the common information node on the connection path between the first insurance product node and the second insurance product node, a second connection weight corresponding to an edge of the second insurance product node connected with the common information node, and a third connection weight corresponding to an edge connected with different common information nodes, a forward similarity of an insurance product corresponding to the first insurance product node and an insurance product corresponding to the second insurance product node includes:

7. The method of claim 6, wherein determining the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path according to the first connection weight, the second connection weight, and the third connection weight on the connection path comprises:

According to the formula:

determining the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path, wherein g is the number of the connection path, value _g For the sub-forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node on the connection path with the number g, M is each common information node on the connection pathThe number of connected edges, m is the number of the edges connected with each commonality information node on the connection path, value _m And the connection weight corresponding to the edge numbered m on the connection path is obtained.

8. The method of claim 7, wherein determining the forward similarity of the insurance product corresponding to the first insurance product node and the insurance product corresponding to the second insurance product node according to the sub-forward similarities on the connection paths comprises:

according to the formula:

9. An insurance product similarity determining device, characterized by comprising: a first difference information node determining unit, a second difference information node determining unit, and an insurance product similarity determining unit,

10. The apparatus of claim 9, wherein the insurance product similarity determining unit includes: a first negative similarity determination subunit, a second negative similarity determination subunit, and an insurance product similarity determination subunit,