CN111951057A

CN111951057A - Advertisement recommendation method and system based on Ether house intelligent contract platform

Info

Publication number: CN111951057A
Application number: CN202010848323.1A
Authority: CN
Inventors: 吴晓晓; 陈诗辉
Original assignee: Shenzhen University
Current assignee: Shenzhen University
Priority date: 2020-08-21
Filing date: 2020-08-21
Publication date: 2020-11-17

Abstract

The invention discloses an advertisement recommendation method and system based on an Ether house intelligent contract platform. The method comprises the following steps: establishing an Etheng binary social graph according to the attention information of the external user account node to the contract account node; for the binary social graph, performing community discovery based on a low-rank graph signal, finding an external user account group with the same interest on a contract account node as a user clustering result, and finding a similar intelligent contract node based on the attention information of the external user account to the contract account as a token clustering result; recommending token information on an Etherhouse based on the user clustering result and the token clustering result. According to the invention, through a community discovery method, the interest communities in the Ethernet block chain network are mined, and a more accurate user cluster is provided for advertisement recommendation; through an ether house block chain platform, the advertisement information distribution under the transfer transaction mode is realized, and the purpose of advertisement recommendation is achieved.

Description

Advertisement recommendation method and system based on Ether house intelligent contract platform

Technical Field

The invention relates to the technical field of block chain application, in particular to an advertisement recommendation method and system based on an Ether house intelligent contract platform.

Background

Intelligent contracts in etherhouses are a well-defined computer protocol that can be organized or created contract code to ensure consistent and enforced coordination among network participants. In the ether house, the user directly transacts via the ether house smart contracts, in effect forming a social network in a particular way. Data in social networks is often very valuable. In a traditional centralized social network, data is stored in a data center of a platform, and the platform usually monopolizes all data and mines user preference information to obtain benefits. However, the blockchain is decentralized, and all users in the blockchain network share data to form a social network. In general, each user on the network has full access to the data. Therefore, data mining of blockchain networks is more valuable than traditional centralized networks because each user can develop its own independent application for distribution to the blockchain.

Traditional social networks are centralized structures that are created by users to create content. Interaction among users is realized through a centralized social network, and a service provider grasps social data generated by the users. However, such a centralized structure has a great risk to the privacy protection of the user. Under certain conditions, the hidden trouble may cause adverse consequences such as information leakage. In addition, in the aspect of traditional internet advertisement, due to a centralized operation mode, the problems of advertisement fraud, inaccurate putting, poor user experience and the like exist. In the block chain network, as a data provider, the block chain data is used for advertisement recommendation, so that the block chain network has a wide application prospect.

Currently, advertisement recommendation is mainly divided into an online mode and an offline mode. The online mode mainly includes the first token Issuing (ICO) and the air drop; the offline mode mainly comprises a road radio, a white paper release and code update. The first token issuance is also called the first token issuance and the block chain crowd funding, and is a way of combining the usage right and the cryptocurrency into one by using the block chain so as to finance the items of developing, maintaining and exchanging related products or services. The air drop in the blockchain field is a market strategy for distributing digital currency to some existing digital currencies in a large scale through blockchain technology, and the air drop in the blockchain field is also regarded as a market strategy for improving the influence of product concepts. The publication of white papers is sometimes carried out simultaneously with the road performance, the idea of the road performance is translated from English Roadshow, the method is a security publishing and popularizing mode widely adopted internationally, and the method is mainly a means for introducing own block chain items and carrying out important recommendation and propaganda. Code updates are typically made on the GitHub web site to indicate the authenticity of the blockchain item.

ICOs, as an online form, need to work in conjunction with offline advertising recommendations. The road performance and the white paper release are a way to publicize the blockchain project, however, in the process, the cost is high, the audience population is less, and therefore the expected revenue of the advertisement cannot be guaranteed; the code updating can provide the authenticity of the project, but requires investors to completely rely on the code of the project to judge the value of the project, and the method can only spread the advertisement information to users with code bases, thereby reducing the service crowd of the advertisement; the airdrop is an online mode, and the current implementation method is to randomly select users to distribute token (advertisement information), which obviously has randomness and makes recommendation of advertisements inaccurate.

In a word, in the existing block chain project advertisement recommendation, the way of showing and issuing the white paper is high in cost, and the audience population is less; the advertising mode of updating the code can serve unique crowds and cannot meet the public demand; the airdrop method is random and the advertisement is not accurate.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides an advertisement recommendation method and system based on an Etheng intelligent contract platform, which is a new technical scheme related to online advertisement recommendation.

According to the first aspect of the invention, an advertisement recommendation method based on an Etherhouse intelligent contract platform is provided. The method comprises the following steps:

establishing an Etheng binary social graph according to the attention information of the external user account node to the contract account node;

for the binary social graph, performing community discovery based on a low-rank graph signal, finding an external user account group with the same interest on a contract account node as a user clustering result, and finding a similar intelligent contract node based on the attention information of the external user account to the contract account as a token clustering result;

recommending token information on an Etherhouse based on the user clustering result and the token clustering result.

According to a second aspect of the invention, an advertisement recommendation system based on an Etherhouse intelligent contract platform is provided. The system comprises:

a data acquisition module: the method comprises the steps that the token holding amount of a user is obtained based on real token transaction data in an Ethernet workshop network, and then a binary social graph of the Ethernet workshop is constructed, wherein the binary social graph is used for representing the attention information of an external user account node to a contract account node;

a community discovery module: the social graph service system is used for executing community discovery based on low-rank graph signals for the binary social graph, finding external user account groups with the same interest on contract account nodes as user clustering results, and finding similar intelligent contract nodes as token clustering results based on the attention information of the external user accounts to the contract accounts;

the advertisement recommending module: for recommending token information on an Etherhouse based on the user clustering result and the token clustering result.

Compared with the prior art, the method has the advantages that the interest communities in the ether house block chain network are mined by the community detection method, so that a more accurate user cluster is provided for advertisement recommendation; advertisement information distribution under the form of transfer transaction is realized through an Ether house block chain platform; through the intelligent contract of the Ether house block chain, the advertisement distribution under the intelligent contract transaction form is realized, and then the purpose of accurately recommending the advertisement is realized.

Other features of the present invention and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description, serve to explain the principles of the invention.

FIG. 1 is a flow diagram of a method for advertisement recommendation based on an Etherhouse intelligent contract platform, according to one embodiment of the present invention;

FIG. 2 is a data broadcast and consensus diagram for an Etherhouse blockchain for one embodiment of the present invention;

FIG. 3 is a schematic diagram of an account generation process according to one embodiment of the invention;

FIG. 4 is a schematic illustration of a transaction type of an Etherhouse of one embodiment of the present invention;

FIG. 5 is a bipartite social graph according to an embodiment of the invention;

FIG. 6 is a diagram illustrating a community discovery method based on low rank map signals according to an embodiment of the present invention;

FIG. 7 is a diagram illustrating a community discovery method based on low rank map signals according to another embodiment of the present invention;

FIG. 8 is a schematic illustration of an Etherhouse transaction type in accordance with one embodiment of the present invention;

FIG. 9 is a schematic diagram of an advertisement recommendation process based on the Etherhouse intelligent contract platform according to an embodiment of the present invention;

FIG. 10 is a schematic diagram of an advertisement recommendation system based on an Etherhouse intelligent contract platform according to one embodiment of the present invention;

FIG. 11 is an illustration of an implementation interface of an advertisement recommendation system based on an Etherhouse intelligent contract platform according to an embodiment of the present invention;

in the figure, Transaction-Transaction; transaction value-Transaction value; InputData-input data; CA node-contract account node; EoA node-external user account node; EoA lead node-external user account leader node.

Detailed Description

Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.

In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.

Referring to fig. 1, in brief, the advertisement delivery method based on the intelligent contract platform of the ethernet workshop provided by the invention includes: step S110, establishing an Etheng binary social graph according to the attention information of the external user account node to the contract account node; step S120, for the binary social graph, executing community discovery based on a low-rank graph signal, finding an external user account group with the same interest on a contract account node as a user clustering result, and finding a similar intelligent contract node based on the attention information of the external user account to the contract account as a token clustering result; and step S130, recommending token information on an Etherhouse based on the user clustering result and the token clustering result. Namely, the invention firstly combines the community detection method to find the user interest community of the Ethernet house network; and then based on the intelligent contract of the Ethermen, advertisement recommendation is realized on the block chain platform. For ease of understanding, examples of ethernet house network, ethernet house network-based community discovery, and ethernet house chaining advertisement placement will be described separately below.

First, about ether mill network

An etherhouse is a distributed platform that runs intelligent contracts. An intelligent contract is an application running on an etherhouse virtual machine that exists in the network as a special account. There are two kinds of accounts in the ether mill, external user account and intelligent contract account, and external user account includes: an address and balance; the intelligent contract account comprises: address, balance, a status, code. It should be noted that there is an Input field in both transactions. If the transaction calls the EtherFang intelligent contract, the transaction will contain a function name and parameters to indicate how to call the intelligent contract. Fig. 2 illustrates the basic process of data broadcasting and consensus of the ether house block chain: user A initiates a transaction to user B; the user A packs the transaction information and broadcasts the transaction information to other nodes of the whole network; other users (including miners) verify the transaction, if the transaction information is correct, the user stores the transaction information into a local transaction pool (the transaction does not represent the consensus of the whole network at the moment); the spacious workers can take the transaction information, can also verify the transaction, and can carry out 'mining' (a method for competing for the accounting right of block data, such as POS, POW and the like), and the miners who succeed in 'mining' can pack a part of transaction information into blocks, assemble the transaction information into block information and then broadcast the block information; and verifying the user passing the block information, and linking the block information with the tail part of the original block data to form a block chain.

There are two types of etherhouse accounts: one is an external account, controlled by a key; the other is a contract account, controlled by the code of the intelligent contract. The external account corresponds to a bank account, and the contract account is an account having a specific function, i.e., an account that invokes an intelligent contract. The generation process of the secret key is shown in fig. 3, the elliptic curve encryption algorithm ECDSA-secp256k1 is used for mapping the private key to generate a public key, and only one public key can be mapped out by one private key; and (3) using a hash algorithm Keccak-256 to hash the public key, converting the public key into 32 bytes, and taking the last 20 bytes as an account address. The external accounts of the etherhouses are transaction-initiable, while the smart contract accounts can only be called, and the transaction types of the etherhouses network are two types, as shown in fig. 4, labeled transaction type 1(i.transaction) and transaction type 2(ii.transaction), corresponding to the two accounts of the etherhouses.

For the intelligent contracts, users can realize own functions on the block chain by writing the intelligent contracts. In connection with the type of transaction of the ether house, it can be found that: transaction type 1 transfers are tai-chi; while transaction type 2 provides an "instruction" and the "Input" field provides the parameters of the "instruction". It should be noted here that there are "Input" fields for both transaction types, this field for transaction type 1 is normally empty, and the field for transaction type 2 is an "instruction" parameter.

Community detection for Etherhouse networks

As a blockchain network, etherhouses differ from bitcoins in many ways. In particular, the etherhouse not only provides a platform for ETH token transactions, but also enables users to publish distributed applications by building intelligent contracts. Each etherhouse user generates a pair of asymmetrically encrypted public and private keys for joining the network. Each public key may be considered a node in the etherhouse social network.

The EOA (account owned by the external user) transaction address and the CA (contract account) transaction address are two transaction addresses in the ethernet arcade, and both have a unique hash address. Thus, according to the properties of the Etherhouse, in one embodiment, the social network of the Etherhouse is defined as a binary social graph, see the left portion of FIG. 5, with the EOA nodes and CA nodes placed on both sides of the graph. Each EOA node will focus on a different CA node, e.g., on the left side of fig. 5, assuming there are N EOA nodes and M CA nodes in the binary social graph. For each EOA node i ∈ (1, 2.. cndot., N), x is defined_i＝[x_i ¹,x_i ²,...,x_i ^j,...,x_i ^M]Information of interest on all CA nodes for EOA node i, where a CA node may be any intelligent contract in an ethernet house. A typical example of a CA node is a token contract created by an ICO (first token issue) event, e.g., x in a token contract_i ^jMay be the balance of user i on token j. In one embodiment, an EtherFang binary social graph is created with token contracts as CA nodes and community discovery is performed on the binary social graph, grouping all EOA nodes.

A variety of methods may be employed to group or cluster the EOA nodes. For example, EOA nodes are clustered using a Community detection algorithm based on low rank graph signals (see "Community detection from low-rank orientations of a graph filter", Wai H T, Segarra S, Ozdaglar A E, et al, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE,2018: 4044-. In particular, assume that all EoA nodes form a low rank logical sub-graph, where some EoA leader nodes will affect the interest of other EoA nodes in CA nodes. Based on this idea, partitioning the EoA node set of the EtherFangdichotomous social graph into subsets with high edge density can be done by observing the output covariance matrix of the low rank graph signals on EoA nodes and applying a spectral clustering algorithm to the covariance matrix. When the community information of the EoA node is acquired, effective strategies can be proposed to put advertisements on the link in the Etherhouse.

In particular, the community discovery problem is considered to be a graph signal processing problem, where the inputs to the graph signal are information of EOA nodes in the social network. On the right side of FIG. 5 are EOA leader nodes, which are defined as

z passing through a filter

Generating output signals on all EOA nodes

Wherein, V and ^ are from the SVD decomposition of S, h_lRepresenting the coefficients of the filter.

The graph signal observed at the EOA node may be represented as:

where t denotes the CA node t, y^tSampled signal, w, representing the "attention" of the EOA node on the CA node t^t～(0,σ²I) Means that the noise signal has a mean value of 0 and a variance of σ²I is a full rank matrix. Assuming an excitation signal

Is zero mean.

To further discuss the structure of the figure, let:

and R is less than N. Due to the input signal z^tFrom the R EOA nodes, the stimulus signal is therefore limited to the R mode.

Further, the signal may be based on observed signals

Obtaining a sample covariance matrix

Then, based on the clustering algorithm in table 1 below, the community discovery problem in S is solved. In fact, the covariance matrix:

the underlying community structure of the social graph may be embodied. For example, a covariance matrix C is calculated_xEstimation of (2):

is C_xIn the above-described method, the estimation of (c),

can be obtained by observing the graph signals of a number of examples t. For example, in the EtherFangdichotomy social graph, each t is regarded as a CA node in the dichotomy graph, and the graph signal is obtained by observing the attention of the EOA node to the CA node

And use of y^tTo detect a community of EOA nodes. In the typical case referred to herein, y^tMay be the amount of different users holding on the token t.

Graph-based signal

The algorithm of table 1 is applied to detect communities of EOA nodes. The abstract of the algorithm model is shown in fig. 6. The numerical analysis will be described later.

Table 1: community discovery algorithm based on low-rank graph signals

Based on the above method, it can be used to find the community of users, i.e. find the EOA node community, which is the most intuitive idea. In another embodiment, this process may be reversed for the EtherFang's CA nodes to cluster the CA nodes through the user's subscription. Referring to FIG. 7, the EOA node and the CA node swap positions and observe the pattern signal at the CA node. Obviously, this signal represents the interest of all EOA nodes for a particular CA node. This is equivalent to replacing in the algorithm

And thereinThe same clustering method is performed. The relationship between the EtherFang network-based user clustering and token clustering is similar to the relationship between user-based collaborative filtering and content-based collaborative filtering, so in conjunction with correlation algorithms, user clustering and token clustering can provide a basis for implementing on-chain advertising.

After obtaining the user clustering result and the token clustering result through community detection, how to put the on-chain advertisement using the clustering result will be discussed hereinafter.

There are two types of transactions in an etherhouse, one is a transaction that creates a smart contract; another is a message-invoked transaction such as transferring ethernet coins or invoking a function of a smart contract, etc. An Etherhouse transaction may contain fields, where different types of transactions may contain fields in common, such as "gasLimit" and "gasTracice", among others; in addition, different fields may be included, based primarily on the purpose of the transaction, e.g., a transaction that creates a smart contract may include an "init" field to indicate the code of the smart contract (which is a description of the function); the transaction of the message call class contains an "input" field to attach additional information, such as parameters required to call a function in the smart contract or information required to be exposed. FIG. 8 shows two transactions of the information call class in the EtherFang blockchain: transaction 1 is a transaction of an ETH token (left of figure 8) and transaction 2 is a smart contract transaction (right of figure 8). Transaction 1 is for an ETH token transfer, while transaction 2 is a transaction that invokes a smart contract. It should be noted that for both transactions, the script has an "input" field, in which parameters can be set for performing corresponding operations in the ethernet house or some information strings to be displayed, so that the chain advertisement can be realized through the "input" field of the ethernet house transaction.

It should be further noted that, in terms of the accuracy of the on-chain advertising, the community discovery algorithm of the low rank graph signal can help find EOA user groups (user clusters) with the same interest on the CA nodes, and find similar CA nodes on the EOA nodes. This provides accurate users and accurate tokens for on-chain advertising. FIG. 9 shows a flow chart of a recommendation strategy, comprising: acquiring the holding amount information of the token of a user through historical data of an ether house; then clustering based on users and clustering based on tokens, wherein the clustering result is a user cluster and a token cluster with 'similarity'; and finally, recommending token information. For example, the recommendation may have three strategies, recommendation based on user clustering results, recommendation based on token results, and recommendation based on a combination of users and tokens. The recommendation based on the user clustering result mainly considers that tokens concerned by users in the cluster are similar after the user clustering is carried out, and then information of other tokens which are not concerned by some users in the cluster is recommended; the recommendation based on the token clustering result mainly considers the similarity of tokens, wherein the similarity refers to similarity of concerned people of the tokens but not similarity of characteristics of the tokens, and then the tokens in a community are recommended to users according to the community to which the tokens concerned by the users belong; the last is based on the recommendation of the user and the token, first find the intersection of the recommended token list of recommendation policy one and the recommended token list of recommendation policy two, and then recommend the information of the intersection token.

Third, example of advertisement placement on Ether house chain

The implementation of the on-chain ad is to add ad information to the "input" field of the deal when the deal is sent (see fig. 8). There are two main approaches, one is to send a small amount of ethernet (which may be 0) to the target user and attach a referral message in the "input" field in this ethernet transfer transaction. This is easy to implement, while sending messages costs a little and advertisement recommendations can only be done in a one-to-one manner. Another approach is similar to so-called "airdrop", that is, some new ICO promoters distribute their partial tokens to the community free of charge to promote their ICO project. This may be accomplished in a group message fashion via an intelligent contract. By this method, although additional ETH tokens are not consumed except for gas (ether house fee), it must be negotiated with the wallet company to register its tokens in the list of target user wallets. Otherwise the user will not see the newly added token transaction and the advertising messages in the transaction. This design differs from the original "airdrop" in that the first step in this design is to mine potential users through community detection and will take full advantage of the "input" field to send advertising messages.

In a preferred embodiment, a field for advertising information is added to the intelligent contract for tokens to introduce item information. The method can realize batch sending only by paying the cost of one transaction. But need to negotiate with the wallet company of the token to register the token in the wallet. Otherwise, the user will not see the newly added advertising information. It can be determined that the design of the present invention is more efficient than the original "air drop" because potential users have been located by community detection and the fields of advertising information have been added to the contract.

In one embodiment, referring to fig. 10, the provided advertisement recommendation system includes: the system comprises a data module (or called a data acquisition module), a community discovery module and an advertisement recommendation module.

And the data module is used for acquiring the real token transaction data in the Ethernet workshop network. Specifically, the "user-token" data may be filtered out through three steps. In a first step, tokens up to 20 in market value were picked from the website "https:// etherscan. io" for etherhouse transaction data prior to 7/1/2019, and the user address 100 before the hold of each token was recorded. In the second step, the transaction to which the user address relates is extracted and information is recorded for all token currencies they currently own (not just the top 20 token currencies mentioned above). By limiting the token currency of interest to the user to no less than 20, information was obtained for the 1837 token currencies of interest to 141 users. In the third step, token data with less than 60 tokens to be focused on is deleted, so that 21 tokens are focused on by this part of users, in order to reduce the amount of data and make the study more targeted. Finally, the information of the holding amount of 21 tokens by the 141 users is retained and recorded as a matrix

Element a in the matrix_ijIs represented byThe holding amount of token j by user i (each column of the A matrix is y)^t). Prior to clustering, the elements of the a matrix will be multiplied by the unit price of the 7/1/7 th ETH in 2019 to unify the absolute value in the matrix.

The community detection module is used for clustering users based on token data of the Ethermen. Specifically, in the dataset, there are 141 users holding 1837 tokens, and the matrix

Too many 0's in the sequence will affect the clustering effect, and further processing is required to the data, so that the number of the concerned users of the token is limited to be not less than 60 in the experiment. The processed data is the holding amount information of 21 tokens by 141 users. This data constitutes a binary social graph of the Etherns, corresponding to the "user-token" data matrix

Wherein A is_ijRepresenting the total holding of token j by user i.

In user clustering, graph signals

Represents the holding amount of the token T (T1, 2.., T) by all users, and is regarded as a feature of the group user. Meanwhile, in order to perform clustering of tokens, redefinition is required

The user is considered a feature of the token. Of the 1837 tokens, some users hold 0, so tokens not held by any user are deleted, 1811 tokens remain, and a new bipartite graph is obtained as a "token-user" matrix

Similarly, by calculating a sample covariance matrix

Then applying the algorithm pairTokens are clustered.

And the advertisement recommending module is used for realizing the advertisement recommending on the ether house chain. For example, existing development software, such as MetaMask, MyEtherWallet, and Remix, may be used. The basic platform of the test experiment is the Ropsten test network of the Ethern. Ropsten is an etherhouse testing environment used to quickly develop intelligent contracts or conduct transactions.

The specific implementation details are as follows: the account is first logged in (as in fig. 11) by using a wallet (MetaMask and MyEtherWallet) and the transaction is constructed and advertising information is added using the MyEtherWallet module (https:// www.myetherwallet.com /). In an advertising implementation that requires intelligent contract participation, Remix modules (https:// Remix. ethereum. org /) are used to deploy related contracts, where the trade address is the intelligent contract that needs to be modified for publication. Finally, the in-chain advertising information can be viewed at the website (https:// etherscan. io) according to the relevant links. The specific usage and verification process for the ethernet network will not be described herein. Through verification, the advertisement recommendation method and the advertisement recommendation system provided by the invention are more accurate, and the benefits brought by the method are more stable; compared with the prior art, the method has the advantages that the advertisement recommendation is carried out in the form of the intelligent contract, the cost is low, the field is added for carrying out the advertisement recommendation, and the recommendation effect is improved.

In conclusion, potential users are searched through a community detection algorithm before advertisement recommendation, and the user community of the Etheng is effectively mined based on community discovery under a node connection low-rank matrix; and then, advertisement recommendation is carried out in each community, so that the advertisement recommendation accuracy is improved. In addition, the invention improves the characteristic of inaccurate advertisement caused by randomness in the traditional air-drop method, combines the result of community detection, and can meet the requirements of the public through advertisement distribution in the form of transfer transaction and intelligent contract transaction, and reduce the advertisement cost.

The present invention may be a system, method and/or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied therewith for causing a processor to implement various aspects of the present invention.

The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.

The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.

The computer program instructions for carrying out operations of the present invention may be assembler instructions, Instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present invention are implemented by personalizing an electronic circuit, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA), with state information of computer-readable program instructions, which can execute the computer-readable program instructions.

Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.

These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, by software, and by a combination of software and hardware are equivalent.

Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the invention is defined by the appended claims.

Claims

1. An advertisement recommendation method based on an Ether house intelligent contract platform comprises the following steps:

2. The method of claim 1, wherein performing low rank graph signal based community discovery comprises:

regarding the community discovery problem as a graph signal processing problem, inputting a graph signal as information of an external user account node in the social network, and defining a leader node z of the external user account, wherein the leader node passes through a filter

The output signal generated at the external user account node is represented as:

the graphical signals observed at the external user account node are represented as:

y^t＝x^t+w^t

where t represents a contract account node t, y^tSampled signal, w, representing the "attention" of an external user account node on a contract account node t^t～(0,σ²I) Means that the noise signal has a mean value of 0 and a variance of σ²I is a full rank matrix;

based on observed signals

And obtaining a sample covariance matrix to perform clustering, obtaining the expected clustering number, and further obtaining the user clustering result and the token clustering result.

3. The method of claim 2, wherein the signal is based on an observed signal

Obtaining a sample covariance matrix to perform clustering includes

Using observed signals

Computing a covariance matrix

Find out

The eigenvectors corresponding to the first K big eigenvalues form a matrix

Where K is the desired number of clusters;

executing clustering algorithm to obtain K clusters C₁,C₂,..,C_K。

4. The method of claim 1, wherein recommending token information on an ether house based on the user clustering results and the token clustering results comprises: recommending tokens concerned by users of the same cluster; recommending other tokens in the concerned token cluster to users of the same user cluster; recommending other tokens within the cluster from the cluster of tokens of interest to the user.

5. The method of claim 1, wherein recommending token information on an ether house based on the user clustering results and the token clustering results comprises:

when a transaction is sent on the chain of etherhouses, the ethernet currency is sent to the target user and a referral message is appended in the "input" field in the ethernet transfer transaction.

6. The method of claim 1, wherein recommending token information on an ether house based on the user clustering results and the token clustering results comprises

When a transaction is sent on the chain of ether houses, a field of advertisement information is added to the intelligent contract of the token to introduce the item information.

7. The method of claim 1, wherein for the binary social graph, x is defined for each external user account node i e {1,2_i＝[x_i ¹,x_i ²,...,x_i ^j,...,x_i ^M]Information of interest on all contract account nodes for external user account node i, a contract account being a token contract created by a first token issuance event, x in the token contract_i ^jIs the balance of user i on token j and M is the number of contracted account nodes.

8. An advertisement recommendation system based on an Ether house intelligent contract platform comprises:

9. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.