TWI480819B - Peer-to-peer based reputation system and method for social network - Google Patents

Description

Peer-based reputation system and method based on social network

The present invention relates to a peer-to-peer reputation system and method, and more particularly to a peer-to-peer reputation system and method based on a social network.

The reputation system establishes the reputation of the user through the online user history electronic transaction behavior and interaction combined with the trust evaluation. Reputation systems have been widely used in e-commerce systems such as eBay, Amazon, and Yahoo, but have not been applied to today's social networks, such as Facebook, Twitter, and Plurk. This is because the existing reputation system evaluation relies on the results of each electronic transaction, but the behavior of the social network is not a transaction, so the existing reputation system cannot be directly applied to the social network. And the central server-style social networking site requires a lot of computing power to add the reputation system to the social network, which will cause the social network owners to spend a lot of cost.

Therefore, there is a demand, how can make the current social network can have a reputation system, but it will not cause the burden of the central server processing, which is an important direction of current research and development.

The present application relates to a peer-to-peer reputation system and method based on a social network, which distributes the computational burden of the central server to each user so that the burden is not increased without increasing the burden on the central server. The reputation system of the social network.

According to an aspect of the present invention, a peer-to-peer reputation method based on a social network is proposed. The peer-to-peer reputation method based on the social network includes the following steps: a start end transmits a to-be-checked component identification code to a first end by an inquiry message, wherein the first end has a first component identifier and a first a tree structure data, the first tree structure data storing a trust value associated with the first component identification code; the first end determining whether the first tree structure data is associated with the component identifier to be inspected a trust value; and when the first tree structure data stores a trust value associated with the component identifier to be checked, the first end returns a trust value associated with the component identifier to be checked back to the start end.

According to another aspect of the present invention, a peer-to-peer reputation system based on a social network is proposed, including a first device. The first device has a first component identification code and a first tree structure data, the first tree structure data storing a trust value associated with the first component identification code, wherein the first device receives an initial device An inquiry message is transmitted, the inquiry message including a component identification code to be checked. The first device determines whether the first tree structure data has a trust value associated with the component identification code to be inspected, and when the first tree structure data stores a trust value associated with the component identifier to be inspected, The first device transmits back the trust value associated with the component identification code to be checked to the initiating device.

In order to make the above-mentioned contents of the present invention more comprehensible, a preferred embodiment will be described below, and in conjunction with the drawings, a detailed description is as follows:

Although the invention will be fully described with reference to the preferred embodiments of the invention, it should be understood that those skilled in the art can modify the invention described herein while achieving the effect of the invention. . Therefore, it is to be understood that the following description is a broad disclosure of those skilled in the art and is not intended to limit the invention.

Please refer to the first figure, which is a peer-to-peer reputation system 100 according to an embodiment of the invention. In an embodiment, the peer reputation system 100 includes, for example, a start end 110, a first end 120, and a second end 130. The start end 110, the first end 120, and the second end 130 respectively include, for example, a starting tree structure data 111, a first tree structure data 121, and a second tree structure data 131.

In an embodiment, the start end 110, the first end 120, and the second end 130 are, for example, different electronic devices, such as an initiating device, a first device, and a second device. The electronic device referred to herein includes a handheld device. Mobile communication devices, computers, home appliances, etc. All electronic devices having computing and storage capabilities are within the scope of the present invention. However, in another embodiment, the start end 110, the first end 120, and the second end 130 are, for example, virtual identification accounts, and the data can be stored in the same server, and any account that is not the same belongs to the present invention. The scope. In addition, each device has, for example, an identification code, for example, the start end 110 has, for example, a start component identifier, the first end 120 has, for example, a first component identifier, and the second terminal 130 has, for example, A second component identification code.

The tree structure data is, for example, related material for storing electronic devices. In an embodiment, for example, it is used to store relationship data related to the electronic device.

Please refer to the second figure, which illustrates the peer-to-peer reputation method 200 based on the social network, please refer to the first figure at the same time. In step S210, the initiating end 110 transmits a to-be-checked component identification code to the first end 120 by an interrogation message, wherein the first end 120 has a first component identification code and a first tree structure data 121, first The tree structure data 121 stores a trust value associated with the first component identification code.

For example, the originating end 110 is to know the credibility of the component identification code to be checked, and the starting end 110 may first search whether the starting tree structure data 111 has the trust degree associated with the component identification code to be inspected. If the information related to the component identifier to be inspected is not found in the initial tree structure data 111, an inquiry request may be sent to other devices, for example, the inquiry message having the component identifier to be inspected is transmitted to the first terminal 120.

Further, the trust value referred to herein is, for example, a trust value associated with the first component identification code. The trust value is calculated, for example, based on the interaction of the first component identification code with one of the other component identification codes, wherein the interaction is classified by category.

For example, the first component identification code and the other component identification code are registered on the social network of facebook, and the first component identification code can actively send a message to the message wall corresponding to the other component identification code, because the first component identification code This action is actively performed, so that the first component identification code has a considerable trust for the other component identification code system, so the behavior can be interpreted as the first component identification code system trusting other component identification codes. It should be noted that the action actively performed by the first component identification code is interpreted as the trust degree of the first component identification code to other component identification codes, but the trust degree is not equal to the trust of the other component identification codes to the first component identification code. degree. The trust of the other component identification code with respect to the first component identification code depends on the degree of interaction of the other component identification codes "actively" with the first component identification code.

Furthermore, the actions of the two identification codes interacting can be divided into at least a first category and a second category. However, the present invention is not limited thereto, and it is still within the scope of the present invention if it is divided into two or more categories. In an embodiment, the first category has a first weight, the second category has a second weight, and the trust value associated with the first component identifier is multiplied by the first weight according to the magnitude of the interaction of the first category. The magnitude of the interaction with the second category is multiplied by the second weight. The trust value can be calculated as follows:

Trust value = Σ W _x C _{( j , i ) x} (Formula 1);

Where Wx stands for weight and x stands for category. For example, W1 represents the first weight of the first category and W2 represents the second weight of the second category. C(j, i) represents the behavior of the jth component identification code to the i-th component identification code. C(j,i)x represents the behavior of the jth element identification code performing the xth class on the i-th element identification code. It is worth noting that C(j,i) is not equal to C(i,j). With the above calculation formula, the trust value associated with the j-th component identification code can be calculated.

In addition, the first category and the second category herein refer to different behaviors into different categories. For example, the j-th component identification code attaches the photo related to the i-th component identification code to the message wall of the i-th component identification code, and can be regarded as the trust of the j-th component identification code to the i-th component identification code, so the behavior can be The classification is the first category, and the behavior of the first category can be higher, and the behavior representing the first category shows greater trust. If the j-th component identification code is simply a message wall that affixes some news to the i-th component identification code, since the news does not represent significant significance, it can be classified into the second category, and the weight of the behavior of the second category can be Lower.

Furthermore, even if the weight of the second category is low, if the j-th component identification code performs a certain number of second category behaviors on the i-th component identification code, since the trust value is accumulated, the number of times the behavior of the second category is performed It will be accumulated as the trust value, so although the weight of the second category behavior is lower, but the trust value accumulated by the behavior of the second category is also a certain degree, it can also be understood that the j-element identification code has a considerable degree of trust. code.

In step S220, the first end 120 determines whether the first tree structure data 121 has a trust value associated with the component identification code to be checked.

In step S230, when the first tree structure data 121 stores the trust value associated with the component identification code to be checked, the first end 120 returns the trust value associated with the component identification code to be checked back to the start end 110.

The starting end 110 can determine the trust degree of the component identification code to be checked according to the trust value of the first component identifier and the component identifier to be checked. That is to say, the originating terminal 110 can determine the identity of the component to be inspected by considering the degree of trust between the first component identifier and the component identifier to be inspected, regardless of the degree of trust between the first component identifier and the first component identifier. Trust.

In another embodiment, the start terminal 110 determines the component to be checked according to the trust value of the first component identifier and the component identifier to be checked, according to the trust value of the first component identifier and the first component identifier. The trustworthiness of the identification code. For example, please refer to the following trust value calculation formula:

Trust value = Σ W _x C _{( j , i ) x} W _request (Formula 2);

The meanings of Wx, x and C(j, i) are as described above and will not be described here. Wrequest represents the starting end 110, that is, the trust level of the message sender to the j-component identification code. For example, the i-th component identification code may be the component identification code to be checked, and the j-th component identification code is the inquiry message receiving end sent by the originating end. In an embodiment, although the j-th component identification code trusts the i-th component identification code to a considerable extent, if the inquiry message transmitting end does not trust the j-component identification code, for example, the value of Wrequest is relatively low, even if The j-th element identification code substantially trusts the i-th element identification code, and according to the above calculation formula, the starting end still does not trust the i-th element identification code.

In summary, although the starting end 110 does not recognize the component identification code to be inspected, the information provided by the first terminal 120 can be used to initially know whether the words and deeds of the component ID to be checked are trustworthy, so that it is not necessary to With the assistance of the central server of the social network, the trust evaluation of the component identification code to be checked can be obtained, which is one of the advantages of the method proposed in the application.

Returning to the flowchart of the second figure, in step S240, when the first end 120 determines that the first tree structure data 121 does not have a trust value associated with the component identification code to be checked, the first end 120 is in accordance with the query message. The path transmits the challenge message to a second end 130.

In step S250, when the second end 130 determines that the second tree structure data 131 has a trust value associated with the component identification code to be checked, the second end 130 returns to the originating end 110 according to the path in the inquiry message. Although the second terminal 130 searches for the trust value associated with the component identification code to be inspected as an example, the present invention is not limited thereto, and if the second terminal 130 has not found the component to be inspected yet The second terminal 130 can still transmit the inquiry message to other terminals according to the path in the inquiry message, for example, a third end, a fourth end, and the like. It is within the scope of the present invention to search for the trust value associated with the component identification code to be inspected at the nth end.

However, it is worth noting that the inquiry message can set the level to be searched. For example, the inquiry message can be set to be inquired from the first end to the fourth end, that is, four endpoints are inquired, and the inquiry is still not found. The trust value associated with the component identification code can abandon the search to avoid endless search actions and waste system resources.

In an embodiment, the starting end 110 can determine the trust degree of the component identification code to be checked according to the trust value of the starting component identifier and the first component identifier. In this embodiment, the start end 110 ignores the trust value between the start end 110 and the first component identifier, and the start end 110 also ignores the trust value between the second component identifier and the component identifier to be checked. The start end 110 The trust value of the second component identification code to be checked for the component identification code can be directly used as a reference for the trust value of the component identification code to be checked.

In another embodiment, the start end 110 is based on the trust value of the start component identifier and the first component identifier, the trust value of the first component identifier and the second component identifier, and the second component identifier and the to-be-checked The trust value of the component identification code is used to determine the trust degree of the component identification code to be checked.

For example, referring to Equation 2, when the second component identifier searches for the trust value associated with the component identifier to be checked and returns it to the originating end 110, the starting end 110 is not directly used, and the system recalculates The trust value between the start component identifier and the second component identifier, that is, the weight of the Wrequest is calculated and considered. However, since the initial component identification code and the second component identification code are not related to each other, an identification code of a third party is required; in this embodiment, the first component identification code and the initial component are identified. Calculating the Wrequest value by the relationship between the code and the second component identification code, for example, directly multiplying the "the initial component identification code to the trust value of the first component identification code" by "the first component identification code to the second component identification code" Trust value is pushed. Although the weight of the Wrequest is calculated in the above manner, the present invention is not limited thereto, as long as the "trust value of the first component identification code of the initial component identification code" and the "first component identification code pair" are used. It is within the scope of the present invention to derive Wrequest from the trust value of the component identification code.

In this way, the trust value allows the user to judge the true degree of other user messages and the reputation of the user, and reduces the probability of malicious behavior of the malicious user. Furthermore, the search reputation method proposed in the present application improves the traditional flood search method on the same-type network, and reduces redundant search information, thereby increasing the efficiency of the search.

In addition, the trust value is divided among multiple users by the peer-to-peer network. Then when a user is abnormal, the trust value can still be accessed. In this way, the problem that all user trust values cannot be accessed when the central server network structure server is abnormal can be avoided.

In view of the above, the present invention has been disclosed in a preferred embodiment, and is not intended to limit the present invention. A person skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the scope of the invention is defined by the scope of the appended claims.

100. . . Peer-to-peer reputation system

110‧‧‧Starting end

111‧‧‧Starting tree structure data

120‧‧‧ first end

121‧‧‧First tree structure data

130‧‧‧second end

131‧‧‧Second tree structure data

S210~S250‧‧‧ Process steps

The first figure is a peer-to-peer reputation system according to an embodiment of the invention.

The second picture shows the peer-to-peer reputation method based on the social network.

100. . . Peer-to-peer reputation system

110. . . Starting end

111. . . Starting tree structure data

120. . . First end

121. . . First tree structure data

130. . . Second end

131. . . Second tree structure data

Claims (17)

A peer-to-peer reputation method based on a social network, comprising: a start end transmitting a to-be-checked component identification code to a first end by an inquiry message, wherein the first end has a first component identification code and a first a tree structure data, the first tree structure data storing a trust value associated with the first component identifier; the first end determining whether the first tree structure data has a trust associated with the component identifier to be checked a value; when the first tree structure data stores a trust value associated with the component identifier to be checked, the first end returns a trust value associated with the component identifier to be checked back to the start end; and when the first The first end determines that the first tree structure data does not have a trust value associated with the component ID to be checked, and the first end transmits the query message to a second end according to the path in the query message. The method of claim 1, wherein the trust value associated with the first component identifier is calculated based on interaction of the first component identifier with one of the other component identifiers, wherein the interaction Classified by category. The method of claim 2, wherein the interaction is at least divided into a first category and a second category, the first category having a first weight, the second category having a second weight, and the The trust value associated with the first component identifier is obtained by multiplying the magnitude of the interaction of the first category by the magnitude of the interaction of the first weight and the second category by the second weight. The method of claim 1, wherein the starting end is judged according to the first component identifier and the trust value of the component identifier to be checked. Break the trust of the component identification code to be checked. The method of claim 1, wherein the starting end has an initial component identification code, and the starting end is based on the first component identification code and the trust value of the component identifier to be inspected, and the initial component identification The code and the trust value of the first component identifier determine the trust degree of the component identifier to be checked. The method of claim 1, wherein the second end has a second component identification code and a second tree structure data, and the second tree structure data is associated with the second component identification code. a trust value; wherein when the second end determines that the second tree structure data has a trust value associated with the component identifier to be checked, the second end returns to the start end according to the path in the query message. The method of claim 6, wherein the starting end has an initial component identification code, and the starting end determines the component to be inspected according to the initial component identifier and the trust value of the first component identifier The trust of the code. The method of claim 6, wherein the start end has a start component identifier, and the start end is based on the start component identifier and the trust value of the first component identifier, the first component identifier And the trust value of the second component identifier and the trust value of the second component identifier and the identifier of the component to be checked to determine the trust degree of the component identifier to be checked. The method of claim 1, wherein the starting end is different from the first end. A peer-to-peer reputation system based on a social network, comprising: a first device having a first component identification code and a first tree knot Constructing data, the first tree structure data storing a trust value associated with the first component identifier, wherein the first device receives an inquiry message transmitted by an initiating device, the query message including a to-be-checked component identification code; The first device determines whether the first tree structure data has a trust value associated with the component identification code to be inspected, and when the first tree structure data stores a trust value associated with the component identifier to be inspected, The first device returns the trust value associated with the component ID to be checked back to the initiating device, and further includes: a second device; wherein when the first device determines that the tree structure data does not exist and the to-be-checked The component identifier is associated with the trust value, and the first device transmits the query message to the second device according to the path in the query message. The system of claim 10, wherein the trust value associated with the first component identifier is calculated based on interaction of the first component identifier with one of the other component identifiers, wherein the interaction Classified by category. The system of claim 11, wherein the interaction is at least divided into a first category and a second category, the first category having a first weight, the second category having a second weight, and the The trust value associated with the first component identifier is obtained by multiplying the magnitude of the interaction of the first category by the magnitude of the interaction of the first weight and the second category by the second weight. The system of claim 10, wherein the initiating device determines the trust degree of the inspected component identification code according to the first component identification code and the trust value of the inspected component identification code. The system of claim 10, wherein the The device has an initial component identification code, and the initial device determines the trust value according to the first component identifier and the identifier of the component identifier to be checked, and the trust component of the first component identifier and the first component identifier The trustworthiness of the component identification code to be checked. The system of claim 10, wherein the second device has a second component identification code and a second tree structure data, the second tree structure data storage associated with the second component identification code a trust value; wherein when the second device determines that the second tree structure data has a trust value associated with the component identifier to be checked, the second device returns to the initiating device according to the path in the query message. The system of claim 15, wherein the initiating device has an initial component identification code, and the initiating device determines the inquiring based on the initial component identifier and the trust value of the first component identifier The trustworthiness of the component identification code. The system of claim 15, wherein the initiating device has an initial component identification code, and the initial device is based on the initial component identifier and a trust value of the first component identifier, the first component The trust value of the identifier and the second component identifier, and the trust value of the second component identifier and the component identifier to be checked are used to determine the trust degree of the component identifier to be checked.