Intelligent information flow processing method and system based on cross-platform data fusion

Abstract

Embodiments of the present specification relate to the field of information technology, in particular to a method and system for intelligent information flow processing based on cross-platform data fusion. The method comprises the steps of: reading a plurality of preset evaluation indexes, generating a plurality of hypothesis descriptions according to the plurality of evaluation indexes; sending a plurality of hypothesis descriptions associated with a target subject identifier to a plurality of data platforms; receiving a judgment result given by the data platform for each hypothesis description; parsing the hypothesis description and the judgment result, and mapping the evaluation index score of each evaluation index; receiving a new description of the information flow about the target subject, combining a preset risk problem to generate a to-be-verified description; mapping the to-be-verified description to the score requirement of each evaluation index, obtaining the verification result of each evaluation index according to the index score; and generating a verification result of the to-be-verified description according to the verification results of all evaluation indexes.

Description

TECHNICAL FIELD

[0001] Embodiments of the present specification relate to the field of information technology, in particular to an intelligent information flow processing method and system based on cross-platform data fusion. BACKGROUND

[0002] With the rapid development of information technology, various data platforms (such as social media, e-commerce platforms, financial systems, public databases, etc.) have accumulated massive information flows related to specific subjects (such as individuals, enterprises, organizations, etc.). These information flows are fragmented, heterogeneous, and dynamically evolving across different platforms, making it complex and difficult to comprehensively assess the behavior, credit, or risk status of a certain subject. Current information processing methods usually rely on a single data source or static rules for judgment, lacking effective fusion mechanisms for cross-platform data, making it difficult to address challenges in information authenticity verification, risk early warning, and intelligent decision support. Especially in the face of network rumors, false propaganda, credit fraud, and other risk problems, how to quickly and accurately automatically verify newly emerging information has become a technical problem that needs to be solved urgently. Therefore, an intelligent information flow processing method and system capable of cross-platform data fusion is urgently needed. SUMMARY

[0003] Embodiments of the present specification describe an intelligent information flow processing method and system based on cross-platform data fusion.

[0004] In a first aspect, the embodiments of the present specification provide an intelligent information flow processing method based on cross-platform data fusion, comprising the steps of:

[0005] reading a plurality of preset evaluation indicators, and generating a plurality of hypothesis descriptions according to the plurality of evaluation indicators;

[0006] sending the plurality of hypothesis descriptions associated with the target subject identifier to a plurality of data platforms;

[0007] receiving the judgment results given by the data platforms for each hypothesis description;

[0008] analyzing the hypothesis descriptions and judgment results, and mapping them into indicator scores for each evaluation indicator;

[0009] receiving a new description of the information flow about the target subject, and generating a to-be-verified description in combination with a preset risk problem;

[0010] mapping the to-be-verified description into score requirements for each evaluation indicator, and obtaining verification results for each evaluation indicator according to the indicator scores;

[0011] generating a verification result of the to-be-verified description according to the verification results of all evaluation indicators.

[0012] In a second aspect, the embodiments of the present specification provide an intelligent information flow processing system based on cross-platform data fusion, characterized in that it comprises:

[0013] a reading module configured to read a plurality of preset evaluation indexes and generate a plurality of hypothesis descriptions according to the plurality of evaluation indexes;

[0014] a sending module configured to send the plurality of hypothesis descriptions associated with a target subject identifier to a plurality of data platforms;

[0015] a receiving module configured to receive a judgment result given by the data platforms for each hypothesis description;

[0016] an analysis module configured to analyze the hypothesis descriptions and the judgment results and map them into index scores of each evaluation index;

[0017] a generation module configured to receive a new description of an information flow about the target subject and generate a to-be-verified description in combination with a preset risk problem;

[0018] a comparison module configured to map the to-be-verified description into score requirements of each evaluation index and obtain a verification result of each evaluation index according to the index scores;

[0019] a result module configured to generate a verification result of the to-be-verified description according to the verification results of all the evaluation indexes.

[0020] In a third aspect, the embodiments of the present specification provide an electronic device, comprising a processor and a memory;

[0021] The processor is connected to the memory;

[0022] The memory is configured to store executable program codes;

[0023] The processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to execute the method of any one of the above aspects.

[0024] In a fourth aspect, the embodiments of the present specification provide a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the method of any one of the above aspects.

[0025] In a fifth aspect, the embodiments of the present specification provide a computer program product comprising a computer program, wherein the computer program is executed by a processor to implement the method of any one of the above aspects.

[0026] The technical solutions provided by some embodiments of the present specification have at least the following beneficial effects:

[0027] In the embodiments of the present specification, the intelligent information flow processing method and system based on cross-platform data fusion provided in the embodiments of the present specification convert evaluation indexes into verifiable hypothesis descriptions, and actively initiate collaborative verification to multiple independent data platforms, utilize feedback results of multiple data sources, effectively reduce the misjudgment risk caused by a single data source, and improve the judgment accuracy of information related to a target subject. The fragmented information in heterogeneous data platforms can be mapped to quantitative index scores, and semantic-level fusion of cross-platform data is achieved. Through the combination of semantic analysis and preset risk problems, newly emerging information flow is automatically converted into to-be-verified descriptions, and efficient verification is performed based on existing index scores, thereby improving the real-time performance of information processing.

[0028] Other features and advantages of the embodiments of the present specification will be further disclosed in the following specific embodiments and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly explain the technical solutions in the embodiments of the present specification, the drawings required to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present specification, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0030] Figure 1 The intelligent information flow processing schematic diagram provided for the embodiments of the present specification.

[0031] Figure 2 The intelligent information flow processing method flowchart schematic diagram provided for the embodiments of the present specification.

[0032] Figure 3 The to-be-verified description generation method flowchart schematic diagram provided for the embodiments of the present specification.

[0033] Figure 4 The intelligent information flow processing system schematic diagram provided for the embodiments of the present specification.

[0034] Figure 5 The electronic device schematic diagram provided for the embodiments of the present specification. DETAILED DESCRIPTION

[0035] The technical solutions of the embodiments of the present specification will be explained and described below in combination with the drawings of the embodiments of the present specification. However, the following embodiments are only preferred embodiments of the present specification, and are not all. Based on the embodiments in the embodiments, other embodiments obtained by those skilled in the art without creative labor also belong to the protection scope of the present specification.

[0036] The terms "first", "second", "third", and the like in the description and claims of the present specification and the above drawings are used to distinguish different objects, and are not used to describe a particular order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but can optionally include other steps or units not listed, or can optionally include other steps or units inherent to the process, method, product, or device.

[0037] In the following description, the appearance of terms such as "inner", "outer", "upper", "lower", "left", "right", etc. is only for the convenience of describing the embodiments and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present specification.

[0038] The data involved in the present application is information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data comply with relevant laws, regulations and standards of relevant countries and regions.

[0039] Before introducing the technical solutions recorded in the present specification, the application scenarios and related technologies of the technical solutions are introduced.

[0040] Cross-platform data fusion can integrate, correlate and collaboratively analyze information from multiple independent and heterogeneous data sources to achieve a more comprehensive understanding of the same target object or event 34. The data platform 22 can cover government open databases, enterprise internal systems, social networks, Internet of Things devices, financial transaction networks, etc., and their data has significant differences in format, structure, update frequency and access mechanism. The goal of cross-platform data fusion is to break down data silos and extract high-value, high-credibility comprehensive information from scattered data through semantic alignment, logical correlation and model-driven methods. The data from a single platform often has limitations or biases, and the fusion of multi-source data can be mutually verified to reduce the risk of misjudgment. Under the premise of ensuring data security and privacy compliance, cross-platform fusion enables collaborative use of data, promoting the transformation of data resources into data assets. Through the unique subject identifier (such as the unified social credit code, user ID, etc.), the cross-platform data is normalized and associated, solving the problems of homonymy and synonymy.

[0041] The present specification provides a cross-platform data fusion-based intelligent information flow processing method and system, please refer to the attached Figure 1, a plurality of evaluation indicators 11 (such as credit status, behavior compliance, risk level, etc.) are preset to generate a plurality of structured "hypothesis descriptions 12". The hypothesis description 12 is a proposition that can be understood and judged by an external system. For example, for an evaluation indicator 11 "fulfillment ability", the system can generate a hypothesis description 12: "the subject has no contract default record in the past year". Then, the hypothesis description 12 is sent to a plurality of external or internal data platforms 22 (such as business systems, judicial databases, e-commerce platforms, credit agencies, etc.) together with the unique identification of the target subject 21 (such as the unified social credit code of enterprises or user ID). Based on the real data stored locally, the data platform 22 independently judges the received hypothesis and returns "support", "partial support", "not support" or "cannot verify" and the like. Cross-platform data value calling instead of raw data sharing is realized, which not only protects the data privacy and security of each platform, but also realizes the effective cooperation of information.

[0042] After receiving heterogeneous feedback from different platforms, the semantic association between the hypothesis description 12 and the judgment result is analyzed, and it is mapped to the quantitative score of the corresponding evaluation indicator 11. For example, if a plurality of authoritative platforms return "support", the "fulfillment ability" indicator score increases; if there is "not support" feedback, the score is deducted accordingly.

[0043] The present specification first provides an intelligent information flow processing method based on cross-platform data fusion, please refer to the accompanying drawings Figure 2 , comprising the steps of:

[0044] Step S1) reading a plurality of preset evaluation indicators 11, and generating a plurality of hypothesis descriptions 12 according to the plurality of evaluation indicators 11.

[0045] The method of generating a plurality of hypothesis descriptions 12 according to a plurality of evaluation indicators 11 comprises:

[0046] Each evaluation indicator 11 is converted into a verifiable proposition form, which includes a condition description and an expected behavior description;

[0047] A plurality of hypothesis condition descriptions and hypothesis behavior descriptions are generated;

[0048] According to the combination of the hypothesis condition description and the hypothesis behavior description, the hypothesis description 12 is generated.

[0049] A plurality of evaluation indicators 11 are read. The evaluation indicators 11 represent the dimensions for comprehensive evaluation of target subjects such as individuals, enterprises, organizations, etc., such as compliance, credit level, behavior stability and risk exposure, or other evaluation indicators 11 configured according to actual needs.

[0050] Each evaluation index 11 is converted into a verifiable proposition form, including a condition description and a corresponding expected behavior description. The condition description is used to describe a certain specific situation or premise state, usually involving time range, event 34 type, subject attribute or environmental constraints; the expected behavior description defines the specific behavior pattern or state result that the target subject should exhibit under the condition, with clear positive or negative direction. Based on the semantic connotation of the evaluation index 11, a number of sets of hypothetical condition descriptions and hypothetical behavior descriptions are generated.

[0051] For example, read the preset evaluation index 11, such as "financial health", "contract performance ability" and "judicial litigation risk". For the evaluation index 11 of "contract performance ability", it is converted into a verifiable proposition form: if the enterprise has participated in commodity procurement activities in the past 12 months, there should be no record of delayed delivery or default refund. Among them, "the enterprise has participated in commodity procurement activities in the past 12 months" is the condition description, and "no record of delayed delivery or default refund" is the hypothetical behavior description.

[0052] Step S2) Send a number of hypothetical descriptions 12 associated with the target subject identifier to a plurality of data platforms 22.

[0053] After the generation of a number of hypothetical descriptions 12 is completed, the generated a number of hypothetical descriptions 12 are associated with the target subject identifier. The target subject identifier is used as a reference for cross-platform data alignment, adopts a standardized unique coding form (such as a unified social credit code, a user ID, a digital identity identifier), ensures that other data platforms 22 can identify the target subject, and avoids errors caused by naming ambiguity or information redundancy.

[0054] Step S3) Receive the judgment results given by the data platform 22 for each hypothetical description 12.

[0055] The method for the data platform 22 to give the judgment results for each hypothetical description 12 includes:

[0056] The data platform 22 reads its internal data resources according to the target subject identifier to obtain the associated data corresponding to the target subject identifier;

[0057] Calculate the matching degree of the associated data and the hypothetical description 12, and obtain the judgment result according to the matching degree, which is one of support, partial support, non-support or unverifiable.

[0058] After receiving the hypothetical description 12 associated with the target subject identifier, the data platform 22 searches and extracts data in its local data resources according to the target subject identifier. The extracted data includes but is not limited to transaction logs, behavior trajectories, attribute archives, event 34 records or relationship networks, etc., thereby forming a set of associated data related to the hypothetical condition.

[0059] The calculation of the matching degree is not a simple keyword comparison, but is based on semantic analysis, rule engine or lightweight inference model, which analyzes the condition description and expected behavior description in the hypothesis, and compares with the actual data. For example, it is necessary to determine whether the condition is established, whether the behavior occurs, whether the time range overlaps, whether the numerical threshold is met, etc. According to the degree of matching, four types of standardized judgment results are generated, including support, partial support, non-support or unable to verify. Support means that the associated data fully confirms that the expected behavior in the hypothesis description 12 is established under the given condition, and the data is highly consistent with the hypothesis. Partial support means that the data supports the hypothesis to some extent, but there are cases of incomplete information, incomplete behavior or only partial sub-conditions. Non-support means that the data clearly shows that the expected behavior does not occur, or there is a direct conflict with the hypothesis content, and the hypothesis is falsified. Unable to verify means that the platform can identify the target subject, but lacks sufficient associated data, or the data permission is limited, the field is missing, the time range is not consistent, etc., resulting in unable to complete effective comparison.

[0060] For example, in the enterprise credit risk assessment scenario, a number of hypothesis descriptions 12 associated with the unified social credit code of the target enterprise are generated and sent to multiple data platforms 22, including e-commerce platforms, business credit systems, judicial judgment documents databases and third-party credit agencies, etc. Each data platform 22 receives these hypothesis descriptions 12 and makes independent judgments based on its internal data resources. The following takes one of the hypothesis descriptions 12 as an example for illustration:

[0061] The hypothesis description 12 is: “If the enterprise has purchase orders in the past 12 months, there should be no platform arbitration records caused by delayed delivery”. The e-commerce platform queries the historical transaction records of the enterprise as a supplier, extracts the number of orders in the past year, the delivery time and whether arbitration is initiated, and the e-commerce platform finds that the enterprise has 20 purchase orders, of which 3 exist delayed delivery, and the buyer has initiated 2 platform arbitrations. Therefore, the condition is established (there are purchase orders), but the expected behavior is not established (there are arbitration records). The business system retrieves the enterprise annual report and administrative penalty information, and the judicial database retrieves whether the enterprise is involved in contract dispute litigation. The business system does not find relevant penalties, but cannot confirm the specific transaction performance, so it cannot directly verify the hypothesis. The judicial database shows no relevant litigation, but the arbitration behavior does not enter the judicial procedure, so it cannot completely deny the hypothesis. Finally, the e-commerce platform gives the judgment result as non-support, and the business system gives the judgment result as unable to verify. The judgment result given by the judicial database is partial support, that is, no litigation record can only indirectly support, so the result is partial support.

[0062] Step S4) Analyze the hypothesis description 12 and the judgment result, and map it to the index score 23 of each evaluation index 11.

[0063] The method of analyzing the hypothesis description 12 and the judgment result and mapping into the index score 23 of each evaluation index 11 includes:

[0064] Calculating the correlation degree of the hypothesis description 12 and each evaluation index 11, and obtaining the correlation weight of the hypothesis description 12 to each evaluation index 11 according to the correlation degree;

[0065] Reading the preset score corresponding to each possible value of the judgment result, and multiplying the preset score by the correlation weight to obtain an adjusted score;

[0066] Superimposing the index score 23 and the adjusted score to update the index score 23.

[0067] The hypothesis description 12 is derived from one or more evaluation indexes 11, and due to semantic expansion and logical generalization, a single hypothesis description 12 can involve the intersection of multiple evaluation indexes 11. Therefore, it is necessary to first calculate the correlation degree between the current hypothesis description 12 and each preset evaluation index 11 through semantic analysis. The correlation degree reflects the influence strength or evidence correlation of the hypothesis description 12 on a certain evaluation index 11. For example, a hypothesis about “contract performance record” may be highly correlated with the “performance ability” index (correlation degree 0.9), moderately correlated with the “operational stability” (correlation degree 0.6), and weakly correlated with the “financial health” (correlation degree 0.2). Based on this, the system assigns an association weight to each hypothesis description 12 on each evaluation index 11 as a basis for subsequent score allocation.

[0068] The four possible values of the judgment result are pre-set to correspond to the basic scores. For example, “support” corresponds to a positive incentive score (such as +10), “partial support” corresponds to a weak positive or neutral score (such as +3), “not support” corresponds to a negative penalty score (such as -10), and “unable to verify” can be set to zero or given an uncertainty reduction score according to the platform credibility (such as 0 or -2). Multiply the basic score by its corresponding correlation weight to obtain an adjusted score. Superimpose the index score 23 and the adjusted score to update the index score 23.

[0069] Step S5) receiving a new description of the information flow 31 about the target subject, combining the preset risk problem to generate a to-be-verified description 30.

[0070] The method of receiving a description of the information flow 31 about the target subject, combining the preset risk problem to generate a to-be-verified description 30 includes:

[0071] Performing semantic analysis on the description of the information flow, identifying events 34 or claims 32 related to the target subject 21, and converting the claims 32 into events 34;

[0072] According to the preset risk problem, a plurality of behavior results 33 of the event 34 are generated;

[0073] The event 34 is mapped as a condition description, and the behavior result 33 is mapped as a behavior description;

[0074] According to the condition description and the behavior description, the to-be-verified description 30 is generated.

[0075] When a new information flow 31 description about a target subject is received, key claims 32 contained therein are automatically identified, and a structured to-be-verified description 30 that can be used for cross-platform verification is generated in combination with a preset risk knowledge system. The received information flow description is subjected to deep semantic analysis. Key claims 32 about the target subject 21, that is, declarative content with verification value, are extracted. For example, “a certain enterprise releases a new generation of products”, “a certain user has abnormal fund flow” and the like. The claims 32 are then abstracted as standardized events 34, including elements such as subject, behavior, object and context, as a basis unit for subsequent reasoning. A set of preset risk problem templates are built in, which are derived from domain knowledge base, management rules, historical cases or expert experience, and cover common risk types such as compliance risk, credit risk, fraudulent behavior, reputation crisis and the like.

[0076] For the identified event 34, a related risk problem is matched, and a plurality of behavior results 33 that the event 34 can cause in a risk situation are deduced accordingly. The behavior result 33 is a logical extension of the potential consequences of the event 34, and represents a verifiable phenomenon that should be observed under a specific risk assumption. For example, if the event 34 is “an enterprise releases a new product”, the corresponding risk problem is “whether there is a false advertising risk?”. Figure 3 After obtaining the event 34 and the corresponding behavior result 33, they are converted into two core components of a logical proposition. The original event 34 is mapped as a condition description, which is used to define the premise situation of verification; the deduced behavior result 33 is mapped as a behavior description, which is used to define the specific behavior pattern that should or should not appear under the condition. The condition description and the behavior description are combined according to the preset logical rules to generate one or more structured to-be-verified descriptions 30.

[0077] Step S6) mapping the to-be-verified description 30 into a score requirement of each evaluation index 11, and obtaining a verification result 41 of each evaluation index 11 according to the index score 23.

[0078] The method for mapping the to-be-verified description 30 into a score requirement of each evaluation index 11 and obtaining a verification result 41 of each evaluation index 11 according to the index score 23 comprises:

[0079] Calculate the relevance of the condition description of the to-be-verified description 30 to each evaluation index 11 respectively, and obtain the relevance weight of the to-be-verified description 30 to each evaluation index 11 according to the relevance;

[0080] Obtain the score requirement of each evaluation index 11 according to the relevance weight and the maximum value of the evaluation index 11;

[0081] When the score requirement is lower than the preset threshold, the verification result 41 of the evaluation index 11 is unverifiable;

[0082] When the index score 23 is higher than the preset effective threshold and higher than the score requirement, the verification result 41 of the evaluation index 11 is support;

[0083] When the index score 23 is not higher than the preset effective threshold and higher than the score requirement, the verification result 41 of the evaluation index 11 is partial support;

[0084] When the index score 23 is not higher than the score requirement, the verification result 41 of the evaluation index 11 is not support;

[0085] According to the verification results 41 of all evaluation indexes 11, the method for generating the verification result 41 of the to-be-verified description 30 comprises:

[0086] Count the verification results 41 corresponding to all evaluation indexes 11, and obtain the verification result 41 of the to-be-verified description 30 according to the verification result 41 with a majority.

[0087] The condition description in the to-be-verified description 30 is semantically analyzed, and the relevance between the condition description and each preset evaluation index 11 is calculated. The relevance reflects the importance or evidence contribution of a certain evaluation index 11 to the judgment of whether the claim 32 is true or false under the current to-be-verified situation. For example, the to-be-verified description 30 is “if an enterprise releases a new product, the product should have no major user complaints in the near future”. The condition “release a new product” may be related to “product quality”, “user satisfaction”, “after-sales service capability” and other indexes. Through a semantic matching algorithm (such as based on word vector similarity, knowledge graph path weight or rule engine), the relevance strength of the condition and each index is quantified, and the association weight of the to-be-verified description 30 to each evaluation index 11 is determined accordingly. According to the association weight and the maximum value (such as 100 points) of the corresponding evaluation index 11, the score requirement required for this verification is calculated as: score requirement = association weight × maximum value of evaluation index 11.

[0088] Combine the current index score 23 of the evaluation index 11 associated with the target subject identifier, compare it with the score requirement generated above, and determine the local verification result 41 of each evaluation index 11 according to the preset threshold rule.

[0089] Specifically, when the index score 23 is higher than the preset effective threshold (higher than the effective threshold means that the index has sufficient reliable data support) and higher than the score requirement, it indicates that the index fully supports the to-be-verified claim 32, and the local verification result 41 is support.

[0090] When the index score 23 does not reach the effective threshold (the data basis is weak), but is still higher than the score requirement, it means that there is a certain basis but insufficient evidence, and the local verification result 41 is partial support.

[0091] When the index score 23 is lower than or equal to the score requirement, regardless of whether the effective threshold is reached, it is considered that the index does not support the claim 32, and the local verification result 41 is not support.

[0092] When the score requirement itself is lower than the preset threshold (such as the correlation weight being too low, which is considered to be irrelevant or negligible), it is determined that the index does not participate in verification, and the local verification result 41 is unable to verify.

[0093] Step S7) generating the verification result 41 of the to-be-verified description 30 according to the verification results 41 of all evaluation indexes 11.

[0094] After obtaining the local verification results 41 of all evaluation indexes 11, the system performs global summarization and statistical analysis. By counting the number of support, partial support, not support, and unable to verify, the number-occupying principle is used to determine the final verification result 41 of the to-be-verified description 30. If the verification results 41 of the evaluation indexes 11 are mostly “support”, then the verification result 41 of the to-be-verified description 30 is “support”.

[0095] On the other hand, the present specification provides an intelligent information flow processing system based on cross-platform data fusion, please refer to the attached Figure 4 , comprising:

[0096] The reading module 100 reads a plurality of preset evaluation indexes 11, and generates a plurality of hypothesis descriptions 12 according to the plurality of evaluation indexes 11;

[0097] The sending module 200 sends the plurality of hypothesis descriptions 12 to the plurality of data platforms 22 in association with the target subject identifier;

[0098] The receiving module 300 receives the judgment results given by the data platforms 22 for each hypothesis description 12;

[0099] The analysis module 400 analyzes the hypothesis descriptions 12 and the judgment results, and maps them into the index scores 23 of each evaluation index 11;

[0100] The generating module 500 receives a new description of the information flow 31 about the target subject, and generates a to-be-verified description 30 in combination with a preset risk problem;

[0101] The comparing module 600 maps the to-be-verified description 30 to a score requirement of each evaluation index 11, and obtains a verification result 41 of each evaluation index 11 according to the index score 23;

[0102] The result module 700 generates a verification result 41 of the to-be-verified description 30 according to the verification results 41 of all the evaluation indexes 11.

[0103] Referring to Figure 5 An electronic device provided by an embodiment of the present specification is shown in a structural schematic diagram.

[0104] As Figure 5 The electronic device 1100 can include at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105, and at least one communication bus 1102. The communication bus 1102 can be used to realize the connection and communication of the above-mentioned components. The user interface 1103 can include a key, and the optional user interface can also include a standard wired interface, a wireless interface. The network interface 1104 can include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc. The processor 1101 can include one or more processing cores. The processor 1101 connects various parts in the entire electronic device 1100 through various interfaces and lines, executes various functions of the routing device and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 1105, and calling data stored in the memory 1105. Optionally, the processor 1101 can be realized by at least one of DSP, FPGA, and PLA. The processor 1101 can integrate CPU, GPU, and modem, etc. The CPU is mainly used to process operating systems, user interfaces, and application programs, etc.; the GPU is used to render and draw the content required to be displayed on the display screen; and the modem is used to process wireless communication.

[0105] It can be understood that the above-mentioned modem can also not be integrated into the processor 1101, but be realized by a separate chip.

[0106] The memory 1105 can include a RAM and can also include a ROM. Optionally, the memory 1105 includes a non-transitory computer-readable medium. The memory 1105 can be used for storing instructions, programs, codes, code sets, or instruction sets. The memory 1105 can include a program storage area and a data storage area, wherein the program storage area can store the instructions for implementing the operating system, the instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), the instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store the data involved in the above-mentioned various method embodiments, etc. The memory 1105 can also be at least one storage device located away from the aforementioned processor 1101. The memory 1105, as a computer storage medium, can include an operating system, a network communication module, a user interface module, and an application program. The processor 1101 can be used to invoke the application program stored in the memory 1105 and execute the method in the above-mentioned embodiments.

[0107] The embodiments of the present specification also provide a computer-readable storage medium, which stores instructions, and when the instructions run on a computer or a processor, the computer or the processor executes the steps in the above-mentioned embodiments. The various component modules of the above-mentioned electronic device, if realized in the form of software function units and sold or used as independent products, can be stored in the computer-readable storage medium.

[0108] The embodiments of the present specification also provide a computer program product, which includes a computer program, and the computer program is executed by a processor to realize the steps in the above-mentioned embodiments.

[0109] The technical features in the embodiments and the implementation forms can be combined arbitrarily without conflict.

[0110] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes a plurality of computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present specification are generated. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transmitted by the computer readable storage medium. The computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with a plurality of available media. The available media can be magnetic media (for example, floppy disk, hard disk, magnetic tape), optical media (for example, digital versatile disc (DVD)), or semiconductor media (for example, solid state disk (SSD)) and the like.

[0111] When implemented by hardware or firmware, the foregoing method processes are programmed into hardware circuits to obtain corresponding hardware circuit structures, and the corresponding functions are implemented. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is an integrated circuit whose logic function is determined by the user programming the device. A digital system is "integrated" on a PLD by the designer himself programming, without having to ask the chip manufacturer to design and manufacture a special integrated circuit chip. Moreover, instead of manually making integrated circuit chips, such programming is now mostly implemented by "logic compiler" software, which is similar to the software compiler used when writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL), and there are many HDLs. Those skilled in the art should also understand that only the method processes need to be logically programmed in the above-mentioned several hardware description languages and programmed into integrated circuits to easily obtain hardware circuits that implement the logical method processes.

[0112] The above-described embodiments are merely exemplary diagnostic description rather than limitations on the scope of the present specification. Any modifications, equivalent replacements, and improvements made to the technical solutions of the present specification by those of ordinary skill in the art, without departing from the design spirit of the present specification, shall fall within the protective scope of the claims of the present specification.

Claims

1. A method for intelligent information flow processing based on cross-platform data fusion, characterized in that, Including the following steps: Read several preset evaluation indicators and generate several hypothesis descriptions based on the evaluation indicators; Several hypothetical descriptions are associated with target entity identifiers and sent to multiple data platforms; Receive the judgment results given by the data platform for each hypothesis description; The hypothetical descriptions and judgment results are analyzed and mapped to the indicator scores of each evaluation indicator; Receive a new information stream describing the target entity, and generate a description to be verified based on preset risk questions; The description to be verified is mapped to the scoring requirements of each evaluation indicator, and the verification result of each evaluation indicator is obtained based on the indicator scores. Based on the verification results of all evaluation indicators, generate the verification results of the description to be verified; The method for receiving a new information stream description about the target entity and generating a description to be verified by combining it with preset risk questions includes: Perform semantic analysis on the description of the information flow to identify events or claims involving the target subject, and transform the claims into events; Based on preset risk issues, several behavioral outcomes for the event are generated; The events are mapped to condition descriptions, and the behavioral results are mapped to behavioral descriptions. Based on the condition description and the behavior description, a description to be verified is generated; The method of mapping the description to be verified to the scoring requirements of each evaluation indicator, and obtaining the verification result of each evaluation indicator based on the indicator scores, includes: Calculate the correlation degree between the condition description of the description to be verified and each evaluation indicator, and obtain the correlation weight of the description to be verified to each evaluation indicator based on the correlation degree. Based on the associated weights and the maximum values ​​of the evaluation indicators, the scoring requirements for each evaluation indicator are obtained; When the scoring requirement is lower than the preset threshold, the verification result of the evaluation indicator is that it cannot be verified. When the score of the indicator is higher than the preset effective threshold and higher than the score requirement, the verification result of the evaluation indicator is supported. When the score of the indicator is not higher than the preset effective threshold but higher than the score requirement, the verification result of the evaluation indicator is partially supported. When the score of the indicator is not higher than the score requirement, the verification result of the evaluation indicator is not supported; Based on the verification results of all evaluation indicators, the methods for generating verification results for the description to be verified include: The verification results corresponding to all evaluation indicators are statistically analyzed, and the verification results of the description to be verified are obtained based on the verification results that are dominant in number.

2. The intelligent information flow processing method based on cross-platform data fusion according to claim 1, characterized in that, Methods for generating several hypothesis descriptions based on several evaluation indicators include: Each evaluation metric is transformed into a verifiable proposition, which includes conditional descriptions and expected behavior descriptions. Generate several descriptions of assumptions and assumptions of behavior; A hypothesis description is generated based on the combination of the hypothesis condition description and the hypothesis behavior description.

3. The intelligent information flow processing method based on cross-platform data fusion according to claim 2, characterized in that, The data platform provides the following methods for determining the outcome for each hypothesis description: The data platform reads the internal data resources of the target entity based on the target entity identifier to obtain the associated data corresponding to the target entity identifier; Calculate the matching degree between the associated data and the hypothetical description, and obtain a judgment result based on the matching degree. The judgment result is one of the following: support, partial support, no support, or cannot be verified.

4. The intelligent information flow processing method based on cross-platform data fusion according to any one of claims 1 to 3, characterized in that, The method for parsing the hypothesis description and judgment results, and mapping them to the indicator scores of each evaluation indicator, includes: Calculate the correlation degree between the hypothesis description and each evaluation indicator, and obtain the correlation weight of the hypothesis description to each evaluation indicator based on the correlation degree; Read the preset score corresponding to each possible value of the judgment result, and multiply the preset score by the associated weight to obtain the adjustment score; The indicator score is then overlaid with the adjusted score to update the indicator score.

5. An intelligent information flow processing system based on cross-platform data fusion, characterized in that, include: The reading module reads several preset evaluation indicators and generates several hypothesis descriptions based on the evaluation indicators. The sending module sends several hypothetical descriptions associated with target entity identifiers to multiple data platforms; The receiving module receives the judgment results given by the data platform for each hypothesis description; The parsing module parses the hypothesis description and judgment results, and maps them to the indicator score of each evaluation indicator; The generation module receives a new description of the information flow about the target entity and generates a description to be verified based on preset risk issues. The comparison module maps the description to be verified to the scoring requirements of each evaluation indicator, and obtains the verification result of each evaluation indicator based on the indicator scores. The results module generates the verification results of the description to be verified based on the verification results of all evaluation indicators. The method for receiving a new information stream description about the target entity and generating a description to be verified by combining it with preset risk questions includes: Perform semantic analysis on the description of the information flow to identify events or claims involving the target subject, and transform the claims into events; Based on preset risk issues, several behavioral outcomes for the event are generated; The events are mapped to condition descriptions, and the behavioral results are mapped to behavioral descriptions. Based on the condition description and the behavior description, a description to be verified is generated; The method of mapping the description to be verified to the scoring requirements of each evaluation indicator, and obtaining the verification result of each evaluation indicator based on the indicator scores, includes: Calculate the correlation degree between the condition description of the description to be verified and each evaluation indicator, and obtain the correlation weight of the description to be verified to each evaluation indicator based on the correlation degree. Based on the associated weights and the maximum values ​​of the evaluation indicators, the scoring requirements for each evaluation indicator are obtained; When the scoring requirement is lower than the preset threshold, the verification result of the evaluation indicator is that it cannot be verified. When the score of the indicator is higher than the preset effective threshold and higher than the score requirement, the verification result of the evaluation indicator is supported. When the score of the indicator is not higher than the preset effective threshold but higher than the score requirement, the verification result of the evaluation indicator is partially supported. When the score of the indicator is not higher than the score requirement, the verification result of the evaluation indicator is not supported; Based on the verification results of all evaluation indicators, the methods for generating verification results for the description to be verified include: The verification results corresponding to all evaluation indicators are statistically analyzed, and the verification results of the description to be verified are obtained based on the verification results that are dominant in number.

6. An electronic device, characterized in that, Including the processor and memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code stored in the memory to perform the method as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-4.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-4.

Images