An internship and employment information management system based on blockchain technology

Abstract

This invention relates to the field of management system technology and discloses an internship and employment information management system based on blockchain technology. The system includes: a business monitoring module that acquires business characteristic data from the mentor's end within the collaborative system; a logic inquiry module that generates inquiry instructions with causal relationships based on task node attributes and transmits these instructions to the student's terminal; an environment verification module that simultaneously captures the wireless signal fingerprints of both parties' environments and calculates the overlap when the student provides response credentials; and a credit evaluation module that determines the authenticity score of the internship behavior based on the response time difference and the overlap. This invention establishes a deep binding relationship between business actions and physical evidence through a collaborative mechanism that simultaneously verifies business logic inquiry and physical on-site characteristics. This fundamentally blocks the path to fraud between mentors and students, ensuring the authenticity of internship credit records.

Description

Technical Field

[0001] This invention relates to an internship and employment information management system based on blockchain technology, belonging to the field of management system technology. Background Technology

[0002] Current internship management systems based on distributed ledger technology provide technical support for the integrity of employment information by storing attendance data and evaluation information during the internship process on the blockchain. Utilizing consensus mechanisms, they ensure the immutability of data after it enters the ledger, thus building a preliminary foundation of trust in administrative management processes. However, in multi-party management scenarios involving students, universities, and enterprises, the system faces the fundamental constraint of data source authenticity. Existing technologies rely excessively on the assumption of the authenticity of on-chain data, focusing on evidence management after data is uploaded to the blockchain, lacking logical verification of the data generation process. When a mutually beneficial relationship exists between the evaluation subject and the evaluated object, the system struggles to identify fraudulent behavior that does not occur physically but is highly disguised at the management logic level. This lack of oversight of the business generation process leads to the risk of the blockchain system recording false management data, creating a technical bottleneck that distorts credit evaluation.

[0003] Simply relying on blockchain backend notarization technology is insufficient to guarantee the authenticity of data sources. Existing information management systems, facing complex interpersonal interactions and vested interests, lack in-depth logical verification of the business generation process. For example, Chinese invention patent CN110796458B discloses an information management system that integrates customer collaboration, marketing management, and user behavior analysis modules to achieve one-stop integration and display of business data. It evaluates performance based on quantitative indicators such as user online time and communication frequency. While such systems improve information flow efficiency and resource integration, their core logic remains based on the assumption that input is always true, focusing on data collection and statistics. They lack a dynamic verification mechanism for the causal relationships of business actions, making it difficult for the system to identify situations involving mutually beneficial internships between mentors and students. The practice of both parties agreeing on pre-defined templates to meet superficial assessment indicators such as online duration or number of communications is compliant in business logic but lacks regulatory oversight, making it easy for the system to record false internship records that did not actually occur in the physical space, resulting in distorted credit evaluations. To address this issue, introducing real-time monitoring based on biometrics or 24 / 7 video verification methods would raise privacy compliance risks and impose high computing power and hardware configuration requirements on SMEs, making large-scale deployment difficult in a universal employment management system. Furthermore, relying solely on geofencing or fixed timestamp verification paths is insufficiently adaptable to premeditated templated reporting content, making it difficult to distinguish between false records and genuine business actions based on the causal density of the management process.

[0004] Therefore, how to use the logical pulses generated by enterprise office business processes to trigger flyback interrogation and combine them with the entropy of wireless environment features to achieve seamless physical presence verification, thereby solving the systemic credit anomaly risk in internship management, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: An internship and employment information management system based on blockchain technology, the system comprising: The business monitoring module is used to collect the business actions of the tutor in the collaboration system and extract business feature information including document operation timestamps and task status identifiers. The logical inquiry module is used to generate inquiry instructions that are causally related to the current business process based on the internship task nodes corresponding to the business characteristic information, and send the inquiry instructions to the student terminal through the distributed ledger. The environment verification module is used to simultaneously capture the wireless signal fingerprints of the environment where the tutor's terminal and the student's terminal are located when the student terminal submits response credentials for the interrogation command, and calculate the overlap of the two sets of wireless signal fingerprints. The credit evaluation module is used to obtain the generation time of business characteristic information and the submission time of response vouchers, and calculate the response time difference. Among them, when the response time difference meets the preset delay threshold and the overlap is higher than the preset similarity threshold, the credit evaluation module determines that the internship behavior is triggered by real business. Distributed ledgers are used to associate and store the judgment results of the credit rating module with business characteristic information.

[0006] Preferably, the logical questioning module performs questioning frequency control based on task risk weights; the logical questioning module identifies the risk weights of each node in the internship task and adjusts the frequency of initiating verification questioning requests according to the risk weights; wherein, when the logical questioning module identifies an assessment node whose risk weight exceeds the preset level, it shortens the allowable range of response time difference and the environmental verification module increases the sampling frequency of wireless signal fingerprints to build a dynamic verification mechanism.

[0007] Preferably, the credit rating module executes a judgment logic based on time-series deviation, and the judgment logic follows the following calculation rules: ,in, For response deviation value; For the response time of the student terminal; The time when business characteristic information is generated; The standard processing time required to execute specific business-related actions; when the response deviation value is within the preset confidence interval, the credit evaluation module determines that the internship behavior conforms to the preset business collaboration rules.

[0008] Preferably, the system also includes a third-party collaborative verification module; the third-party collaborative verification module is used to randomly select verification nodes that are not tutor subjects from the enterprise identity identifier database to participate in consensus; the verification nodes cross-compare the business feature data with the response credentials of the student terminal, extending the verification process from two-person verification between tutor and student to a multi-subject random verification mode.

[0009] Preferably, the system also includes a credit weight adjustment module; the credit weight adjustment module is used to dynamically assign evaluation weights to different internship subjects based on historical credit deviation data; the credit evaluation module adjusts the threshold of the response time difference required to determine whether the internship behavior is a real business trigger based on the weight factors output by the credit weight adjustment module.

[0010] Preferably, when comparing the similarity of wireless signal fingerprints, the environment verification module extracts the service set identifier list of surrounding wireless access points and their corresponding signal reception strength vectors, and establishes an overlap model that reflects the degree of overlap in physical space; wherein, the sampling frequency of the environment verification module in capturing wireless signal fingerprints is not less than 10Hz.

[0011] Preferably, the distributed ledger establishes a data channel for internship tasks through a consensus mechanism; the distributed ledger associates and stores each query instruction initiated by the logical query module with the judgment result of the credit evaluation module to achieve the traceability of credit data.

[0012] Preferably, the business monitoring module includes a silent data acquisition unit; the silent data acquisition unit is used to monitor the document processing node information of the tutor end in the office collaboration system without collecting geographical location information, and convert it into business feature information.

[0013] Preferably, after the credit evaluation module determines that the internship behavior is triggered by a real business, it triggers the distributed ledger to perform a credit record update operation. The credit record update operation includes binding the feature summary of the response voucher with the corresponding business evidence identifier on the chain and updating the accumulated credit score of the internship credit file.

[0014] Preferably, the challenge instruction generated by the logic challenge module includes a verification token; the verification token is used to require the student terminal to simultaneously upload a snapshot of the real-time signal characteristics generated by the environment verification module when responding, as an input basis for determining the degree of overlap.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In internship and employment information management, an asynchronous logical inquiry mechanism is adopted. By logically associating the inquiry initiation identifier, response content identifier, and the preceding task identifier of the student's mobile terminal, an evidence package with causal dependency is constructed. This verification path, which is composed of random offsets of management logic, transforms the originally isolated internship data records into causal chains with information entropy, increases the real-time collaborative cost of collusion and fraud by multiple parties, and ensures the self-consistency and credibility of internship evaluation information in the dimension of management logic.

[0016] 2. By adopting a business stress-driven counter-interrogation logic, the non-confidential business flow feature vector generated by corporate mentors on their office terminals is extracted and used as a dynamic parameter to synthesize collaborative confirmation tasks for students' mobile devices in real time. Since the interrogation trigger source comes from unpredictable real business actions of enterprises, it breaks the static data fabrication mode that relies on preset outlines for responses in existing technologies. It requires both parties involved in the fabrication to maintain complex logical collaboration under dynamic business pulses, thereby blocking the path of generating false management records that are detached from real production relations at the management essence level.

[0017] 3. Through the environmental feature entropy synchronous verification mechanism, the non-connected wireless signal characteristics around the enterprise client and the student mobile terminal are obtained at the moment of the query response. The topological overlap of the two sets of environmental hash mirrors is compared to determine the physical coexistence of the two parties. The naturally existing radio electromagnetic wave noise fluctuations in the environment are used as verification anchor points. Without the need to collect privacy data such as geographical coordinates, the possibility of falsifying remote negotiations using remote control tools or instant messaging tools is eliminated, realizing a deep coupling verification of the physical on-site state and the authenticity of business actions. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the closed-loop verification process of internship collaboration that binds business actions and physical empirical evidence in this invention. Figure 2 This is a five-dimensional module architecture diagram of the system that integrates logical challenge and physical verification functions according to the present invention. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. The following embodiments are intended to explain and illustrate the present invention, and are not intended to limit the scope of protection of the present invention.

[0020] This invention provides an internship and employment information management system based on blockchain technology, including an identity mapping module, a task topology configuration unit, an asynchronous logic query engine, an evidence chain weaving module, a distributed ledger layer, and a credit adjudication unit. Each module uses a distributed identity protocol and an asynchronous logic query mechanism to anchor the authenticity of the internship process. The identity mapping module establishes identity identifiers based on asymmetric encryption algorithms for university management terminals, enterprise clients, and student mobile terminals. The task topology configuration unit receives internship task logic chains uploaded by the university management terminal, which contain multiple business nodes with pre-determined dependencies. The asynchronous logic query engine sends management query requests to the enterprise client within the execution cycle of the business nodes. The evidence chain weaving module collects real-time response data from the enterprise client to the query requests and associates it with the task output hash of the student mobile terminal to form an evidence package. The distributed ledger layer stores the evidence package to blockchain nodes through a consensus protocol. The credit adjudication unit calculates the corresponding internship credit weight based on the continuity of the evidence package and the time characteristics of the query response. In internship management application scenarios, there may be collaborative behaviors between the evaluation subject and the evaluated object that are detached from business actions. The system uses the identity mapping module to construct a distributed identity protocol for internship participants. The protocol uses a unique identifier; it calls an asymmetric encryption algorithm to generate a public-private key pair, and encapsulates the public key and descriptor into a standard-compliant digital identity document stored in the distributed ledger layer; when an enterprise mentor connects to the system, the identity mapping module assigns it a globally unique identifier. Subsequent business actions and inquiry responses must be signed electronically using the associated private key, thereby determining the identity of the operating entity and the auditability of its behavior at the starting point of the management process; the identity mapping module executes hardware-level identity binding procedures, calling the trusted execution environment built into the mobile terminal. The generated unique private key signs the identity document, and the device hardware identifier of the mobile terminal is extracted and hashed to generate a device fingerprint. The device fingerprint and the digital identity document are logically interlocked and stored in the distributed ledger layer. The physical uniqueness of the operating entity is determined by verifying the consistency between the signature of subsequent business actions and the device fingerprint.

[0021] To address the technical issue of the lack of correlation between management records and production processes in internship tasks, the system employs a task topology configuration unit and an asynchronous logic query engine. The task topology configuration unit predefines the sequence of business nodes in the internship outline and assigns a risk weight to each node that reflects the importance of the task. The asynchronous logic challenge engine executes business stress-driven counter-challenge logic. This engine obtains non-confidential business flow feature vectors triggered by enterprise mentors through a pre-set business action monitoring interface on the enterprise client. These feature vectors include timestamp summaries of document modifications, schedule change events from office collaboration software, and process status transition identifiers from the task management system. The engine extracts key business semantic points from the business flow feature vectors and combines them with the feature attributes of the current business node in the internship task logic chain to generate dynamic challenge requests with causal relationships in business evolution. The business monitoring module captures document revision event streams through the office collaboration software hook program and extracts metadata containing revision character increments, revision start and end line numbers, and revision timestamps. The logic challenge module retrieves the metadata and uses a weight matrix... Semantic mapping algorithms generate challenge feature vectors; Pre-processing and pre-fixing are performed using word frequency statistics from an industry corpus; the product of keyword weights in the revised text is calculated and normalized to a fixed-length feature code; the challenge instruction combines the feature code with risk weights. Combined with the student terminal collaborative verification task; the interrogation payload is directly mapped from the business output of the tutor terminal. The causal binding causes the pre-made response deviating from the production process to generate a logical deviation alarm due to the inability to match the semantic feature summary. If it is detected that the tutor has updated a document named Project Feasibility Analysis in the collaboration system, the engine generates a collaboration confirmation instruction for the student's mobile terminal, requiring the student to submit the corresponding survey sketch or meeting minutes hash within the preset response time difference. This mechanism uses random pulses generated by the business flow to trigger interrogation, increasing the collaboration cost for both parties in the fraud in a dynamic environment.

[0022] To ensure that students and tutors are in the same physical space, the system is configured with an environment verification module. The moment the enterprise client responds to an inquiry, this module drives both the tutor's and student's mobile terminals to collect fingerprints of surrounding non-connectivity wireless signals; it then extracts the service set identifiers of the surrounding wireless access points. The list and corresponding signal reception strengths are as follows When constructing a wireless signal fingerprint, the vector environment verification module calls the terminal's wireless sensing interface to scan for access point information in the surrounding 2.4GHz frequency band and extracts information including the service set identifier. and received signal strength indication Feature pairs; The unit is dBm, and the value ranges from -100 to -20; for the same Gaussian filtering is applied to multiple sets of data within the sampling window to extract the median intensity feature, suppressing multipath background noise; multiple feature pairs are then processed according to... The lexicographical order forms the signal feature vector; the credit evaluation module calculates the cosine of the angle between the vectors from the tutor and student ends to determine the degree of overlap. ; The value is a dimensionless value, ranging from 0 to 1; the system uses a calibrated similarity threshold. Judgment; Measured overlap Not less than The system determines that the internship collaboration has a physical presence and performs a one-way hash operation on the radio environment characteristics to generate an environment characteristic summary. and The credit adjudication unit calculates the overlap between two sets of environmental feature summaries. If the Hamming distance between the two sets is within a preset similarity threshold, it determines that the two sets are in the same physical field. Frequency capture includes The signal sequences identified by each access point, if they overlap The proportion of the number exceeds If so, it is determined that the two parties physically coexist within a specific office space.

[0023] The system allocates and manages resources through credit adjudication and credit accumulation units, and the asynchronous logic challenge engine adjusts the probability of initiating challenge requests according to a formula. : ,in, The probability of initiating a challenge request; Risk weight for the current business node; Historical credit deviation; The current total load capacity of the system; total system load capacity The initial value is determined based on the processing performance of consensus nodes in the distributed ledger layer and network bandwidth; the maximum transaction throughput per unit time is obtained through stress testing during the system startup phase. The unit of measurement is transactions per second, ranging from 100 to 5000; the credit evaluation module monitors the average feedback time of consensus nodes and identifies fluctuations in processing capacity; when the average feedback time exceeds a preset threshold, it is reduced proportionally. Reflects the system's load status; Participation initiation probability Calculation and dynamic calibration enable the asynchronous logic challenge engine to reduce the challenge frequency under high load, ensuring the continuous execution of core evidence storage services; for low-risk nodes, the credit accumulation unit temporarily stores the response hash sequence that has not yet been uploaded to the chain on the student's mobile device, and when the local credit accumulation value... Reaching the stage risk threshold At that time, the system performs an on-chain anchoring operation of the aggregate hash. At high-risk nodes involving critical project delivery, the density of challenge requests is increased, and the effective time for the evidence chain weaving module to receive response data is shortened simultaneously.

[0024] To break down the closed-loop collaboration between mentors and students, the system introduces a third-party collaborative verification module. This module randomly selects third-party witness nodes from non-mentor entities belonging to the same enterprise identifier as defined by the identity mapping module to participate in credit consensus. When an inquiry occurs, a confirmation task is sent to the witness node, requiring it to report the operational status of a specific office area. The credit adjudication unit obtains the confirmation information from the witness node and compares its correlation with the enterprise client's response data in terms of management logic. The credit evaluation module executes a judgment logic based on time-series deviation, and its calculation rules follow the formula: ,in, For response deviation value; For student response time; The time when business characteristic information is generated; The standard processing time required to execute a business action; when the response deviation value When the internship credit record is within the preset confidence interval and the multi-dimensional witness feedback is consistent, the system updates the internship credit record on the blockchain. This process improves the objectivity of the internship credit record by introducing a third-party variable. The distributed ledger layer executes an asynchronous consensus program based on logical consistency verification. After receiving the evidence package sent by the evidence chain weaving module, each consensus node retrieves the topology state of the corresponding internship task logic chain, verifies whether the hash of the current business action conforms to the preset pre-order dependency relationship of the task topology configuration unit, and triggers the state synchronization instruction between nodes to solidify the internship record into the blockchain data structure when the calculated logical deviation is within the preset business collaboration dynamics distribution interval.

[0025] Example 1: In a specific engineering design internship scenario, a labor cooperation relationship based on an asymmetric encryption algorithm is established between the intern and the evaluated. In this scenario, both parties can easily reach an agreement on pre-made weekly internship reports and attendance records. This leads to a deficiency in adaptability of verification paths relying on geofencing or fixed timestamps when faced with logically consistent fictitious records. The system calls the identity mapping module to assign a unique digital identity to the enterprise mentor. Furthermore, when the tutor performs document modification actions in the collaboration software, the asynchronous logic query engine collects business flow feature vectors containing timestamp summaries, generating dynamic query requests that require students' mobile devices to submit the corresponding survey sketch hash within a preset response time difference, thus transforming isolated records into causal chains driven by real-time business actions.

[0026] The environment verification module collects wireless access point information of the environment where the tutor and student terminals are located at the instant the student provides the response credentials, extracts the signal received strength vector, and obtains an environment feature summary by performing a one-way hash operation on the radio environment features. and The credit adjudication unit calculates the overlap between two sets of summaries. When the service set identifiers overlap, the overlap is determined by the degree of overlap between the two sets of summaries. The proportion exceeds When both parties are determined to be in the same local physical field, the credit adjudication unit adjusts the initiation probability of the inquiry request according to a formula. : ,in, The probability of initiating a challenge request; Risk weight for the current business node; Historical credit deviation; This represents the current total load capacity of the system. This approach concentrates management resources on business nodes with higher risk weights, temporarily stores low-risk response hashes on the student's end through a credit accumulation unit, and performs an on-chain operation of the aggregated hash when the local credit accumulation value reaches the risk threshold.

[0027] Example 2: In a simulated internship management environment based on a distributed cluster architecture, the effectiveness of the system in identifying logically coordinated fraudulent behavior is verified under the condition that the evaluation subject and the evaluated subject reach a non-cooperative game contract; deployed on a private blockchain network containing ten nodes, using a main frequency of not less than Multi-core processor, configuration of student mobile simulation device The device has an operating system. The protocol's wireless network awareness capability, and the sensor sampling frequency is set to... The data acquisition process introduces network latency jitter, and the latency distribution range is set within... to Between, and superimposed signal-to-noise ratio is Background Gaussian noise is used to simulate complex industrial communication environments.

[0028] During system operation, the asynchronous logic challenge engine dynamically adjusts the sampling period. The setting of the sampling period depends on the technical trade-off between data real-time performance and system computational load. When the rate of change of the business flow feature vector captured by the business action monitoring interface increases, in order to avoid aliasing of challenge commands in the time domain and causing causal chain breaks, the sampling period is adjusted to the lower limit of its range. Approaching; under the typical operating conditions of this experiment, risk weights are set. System load capacity After the system completes digital identity activation through the identity mapping module, the task topology configuration unit generates a logical chain containing five serial business nodes. The experiment is divided into a control group and the present invention sample group. The control group adopts the traditional fixed geofence and periodic check-in mode, while the present invention sample group executes the asynchronous logical challenge and physical on-site dual verification procedure. The experiment process involves stress testing by injecting pre-made fake business records. In the core deduction stage, the system detects the behavior pulse of the tutor modifying the technical solution document, and the asynchronous logical challenge engine then generates challenge instructions. The specific test results are shown in Table 1.

[0029] Table 1: Comparison Table of Test Results for Determining the Authenticity of Internship Behavior

[0030] In Table 1, To respond to time difference, This refers to the overlap of environmental features; data shows that, in real business collaboration, the timing logic of the response credentials generated by the student and the business actions of the tutor matches, the response time difference is within the preset correlation delay window, and the overlap of environmental features is high. Higher than The similarity threshold; in the case of logical fraud, because the two parties involved in the fraud need to synchronize information offline to match business semantics, a response time difference occurs. Increase to This triggered the judgment logic based on time-series deviation; the calculation rules for this judgment logic executed by the credit rating module follow the formula: ,in, In response to the deviation value, For student response time, For the time when business characteristic information is generated, The standard processing time required to execute business actions; in remote negotiation fraud scenarios, although both parties shorten the logical alignment time, Down to However, due to geographical differences, the Hamming distance of the environmental feature summaries exceeds the threshold, resulting in a low overlap. Only Based on this, the system determines that the physical presence is missing.

[0031] When environmental characteristics overlap Reduce to Below the critical value, the credit weight exhibits a non-linear decreasing trend; in the boundary state test group, the overlap... The gradient drop in credit weight settlement indicates that the system has high accuracy in distinguishing changes in physical space; the asynchronous logic challenge engine calculates the initiation probability based on the formula. With risk weight The increase shows a positive correlation: ,in, The probability of initiating a challenge request. Risk weight for the current business node Historical credit deviation, This represents the current total load capacity of the system; when from Upgraded to As the query density increases, this step response effectively suppresses non-genuine business operations under high-risk business nodes; the system uses unpredictable business flow logic pulses as credit anchors to eliminate the static disguise of fraudulent behavior in the time domain; the coordinated operation of asynchronous logic query mechanism and wireless signal fingerprint verification realizes the binding of management records and physical evidence, improving the asset authenticity of internship credit data on the blockchain ledger without increasing additional hardware costs.

[0032] Example 3: This example combines Figures 1 to 2 Describe an internship and employment information management system based on blockchain technology, such as... Figure 1 As shown, the business collaboration action is generated by the tutor. After extracting the business features, the business monitoring module collects the business action and timestamp. The logic questioning module generates a causal questioning instruction based on the task node attributes and sends it to the student terminal. The response is triggered when the response certificate is submitted. The environment verification module captures the fingerprint of the wireless signal from both ends and performs the operation of capturing environmental features to transmit the fingerprint overlap data to the credit evaluation module. The credit evaluation module combines the business timestamp to calculate the time difference and fingerprint overlap to generate a certificate authenticity score. Finally, the distributed ledger completes the judgment result and feature association certificate storage.

[0033] like Figure 2 As shown, the functional structure is divided into five core modules. The business logic inquiry module includes risk weight adjustment, causal instruction generation, and business action monitoring functions. The credit evaluation system module covers the dimensions of dynamic credit accumulation, logical consistency, and response timing deviation. The consensus and evidence storage module consists of distributed ledger, third-party witnessing, and evidence chain weaving. The environmental physical verification module is responsible for physical coexistence determination, signal entropy synchronization, and wireless fingerprint collection. The identity and security module integrates asymmetric encryption, TEE hardware binding, and DID identity mapping mechanism.

[0034] Example 4: In an internship management scenario involving remote offline collaboration, the enterprise mentor performs structural revisions on multiple internship output documents in an environment without network coverage. After network access is restored, batch transaction synchronization is performed. Due to the high-density burst characteristics of business flow pulses in the time domain, the asynchronous logic query engine executes a semantic hash feature mapping program based on a sliding window to establish a definite causal relationship. Its processing logic includes retrieving data generated by the enterprise collaboration software. Format document change records and extract operation subject identifiers, paragraph indexes, and incremental text character streams, using a length of [length missing]. A sliding window performs slicing on the incremental text and uses a weight matrix to perform weighted operations on the keyword frequencies within the slices to calculate the paragraph semantic vector. This is achieved by converting the paragraph semantic vector into a fixed-length... A binary digest value is used as the core payload of the challenge instruction. The sliding window length is set to [value]. One character.

[0035] In handling implicit negative event log injection such as response timeouts or rejections, the system establishes an automated default auditing procedure through transaction channels at the distributed ledger layer, with the credit adjudication unit initiating and business node risk weighting. A negatively correlated time-to-live timer is used to define the response period. If the student's mobile device does not return the survey sketch hash or environmental verification fingerprint before the timer value is cleared, the credit adjudication unit generates an abnormal transaction record marked as a missing response in the distributed ledger. This record contains the original hash of the challenge instruction, the expiration timestamp, and a negative credit offset. Negative credit offset The calculation rules follow the formula: ,in, This represents a negative credit offset. Risk weight for the current business node; The response timeout duration is in seconds; to determine the selection criteria for the similarity threshold in the environmental verification module, the system executes a calibration procedure based on radio environment background feature analysis, performed by two adjacent mobile terminals in a standard office scenario. The environmental feature fingerprint was collected in succession to statistically analyze the wireless access point service set identifier. The variation pattern of environmental noise was analyzed by calculating the variance of the signal received strength at each access point in the sample sequence and determining the confidence interval of environmental noise based on the normal distribution model. The system also analyzed the overlap of environmental characteristics at different workstations within the same office. The sample data, the actual measured mean is at to Between these, the lower limit of the similarity threshold is determined based on the distribution pattern of the samples. This value is capable of identifying remote negotiation fraud when the Hamming distance exceeds a preset boundary, while tolerating differences in antenna gain of mobile terminals.

[0036] Example 5: During the pre-deployment debugging phase of the system in the new office environment, the environmental verification module executes a threshold calibration procedure based on the distribution patterns of environmental characteristics. Five sampling points with physical intervals are selected within the target area, driving the standard mobile terminal to... The frequency of execution duration is not less than Wireless signal background scanning to extract the average received signal strength in the field greater than Given a stable set of wireless access points, calculate the average Hamming distance of the wireless signal fingerprint between each sampling point. and its standard deviation The similarity threshold for this specific field is determined according to the formula. : ,in, This is the similarity threshold for the field; The average Hamming distance between sampling points; Standard deviation; To determine the total number of stable wireless access points detected, the deployment and debugging procedure enables the system to adjust decision boundaries based on the fluctuations in radio characteristics across different physical spaces.

[0037] When the system faces situations where the pace of business collaboration varies across different industries, the credit adjudication unit determines the standard processing time for each business node. Execute the historical business stress mapping procedure and retrieve the weights of logical chain nodes related to the current internship task from the enterprise collaboration log. Matching similar business records, analyzing the median time interval between the mentor's document submission action and the feedback action of the collaborating participants. And use it as the standard processing time. The initial input value is determined based on the actual processing time of the most recent 30 closed-loop business nodes during the internship operation cycle. The specific parameter adjustment logic for performing a moving average update follows the formula: ,in, This is the updated standard processing time; This is the standard processing time before the update; This represents the actual processing time of the current business node. For the range of values ​​within to The smoothing coefficient between them makes the timing judgment logic compatible with the business characteristics generated by different enterprise management rhythms.

[0038] Example 6: In the scenario of configuring cross-industry internship tasks, due to the differences in the vocabulary distribution characteristics of office documents in different disciplines, the system executes a semantic weight matrix calibration program based on a reference corpus to retrieve the target discipline's... Using the internship task outline as the basic corpus, text frequency inverse document frequency was used... The algorithm extracts business-identifiable information. Calculate the feature distribution probability of each core technical term across task nodes. The initial weights are determined based on the relevance of each term to the internship objective, and the asynchronous logic query engine is driven to execute at least [number missing] times the preset business flow. In the hash collision test, the mean cosine similarity of the generated paragraph semantic vectors was higher than [a certain value]. Time-fixed weight matrix .

[0039] When the system is applied to enterprise environments with varying requirements for process rigor, the task topology configuration unit executes risk weights based on task entropy values. The normalized calibration procedure breaks down the individual business nodes in the internship syllabus into... The information entropy model is established by calculating the temporal evolution probability of each independent operation item, as shown in the following formula: ,in, The information entropy value of the task node; For operation of sub-items; This represents the probability of the corresponding operation item appearing. The credit adjudication unit will output the information entropy value of the model as the total number of sub-items operated on. Mapping to from to Within the numerical range to determine the risk weight of the corresponding node By executing a period of time before running The silent monitoring task is used to collect the average business flow feature vector under normal office conditions within the enterprise to generate frequency. This frequency is used as a calibration benchmark for adjusting the initial threshold of the survival timer. This pre-debugging procedure maps the decision logic to the business stress intensity of a specific industry.

[0040] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An internship and employment information management system based on blockchain technology, characterized in that, The system includes: The business monitoring module is used to collect the business actions of the tutor in the collaboration system and extract business feature information including document operation timestamps and task status identifiers. The logical inquiry module is used to generate inquiry instructions that are causally related to the current business process based on the internship task nodes corresponding to the business characteristic information, and send the inquiry instructions to the student terminal through the distributed ledger. The environment verification module is used to simultaneously capture the wireless signal fingerprints of the environment where the tutor's terminal and the student's terminal are located when the student terminal submits response credentials for the interrogation command, and calculate the overlap of the two sets of wireless signal fingerprints. The credit evaluation module is used to obtain the generation time of business characteristic information and the submission time of response vouchers, and calculate the response time difference. Among them, when the response time difference meets the preset delay threshold and the overlap is higher than the preset similarity threshold, the credit evaluation module determines that the internship behavior is triggered by real business. Distributed ledgers are used to associate and store the judgment results of the credit rating module with business characteristic information.

2. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, The logical inquiry module performs inquiry frequency control based on task risk weights; the logical inquiry module identifies the risk weights of each node in the internship task and adjusts the frequency of verification inquiry requests according to the risk weights; when the logical inquiry module identifies an assessment node whose risk weight exceeds the preset level, it shortens the allowable range of response time difference and the environmental verification module increases the sampling frequency of wireless signal fingerprints to build a dynamic verification mechanism.

3. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, The credit rating module executes judgment logic based on time-series deviation, and the judgment logic follows the calculation rules as follows: ,in, For response deviation value; For the response time of the student terminal; The time when business characteristic information is generated; The standard processing time required to execute specific business-related actions; when the response deviation value is within the preset confidence interval, the credit evaluation module determines that the internship behavior conforms to the preset business collaboration rules.

4. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, The system also includes a third-party collaborative verification module; The third-party collaborative verification module is used to randomly select verification nodes that are not tutor entities from the enterprise identity database to participate in consensus; the verification nodes cross-compare business feature data with the response credentials of the student terminal, extending the verification process from a two-person verification between tutor and student to a multi-subject random verification mode.

5. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, The system also includes a credit weight adjustment module; the credit weight adjustment module is used to dynamically assign evaluation weights to different internship subjects based on historical credit deviation data; the credit evaluation module adjusts the threshold of response time difference required to determine whether the internship behavior is a real business trigger based on the weight factors output by the credit weight adjustment module.

6. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, When comparing the similarity of wireless signal fingerprints, the environment verification module extracts the service set identifier list of surrounding wireless access points and their corresponding signal received strength vectors, and establishes an overlap model that reflects the degree of overlap in physical space; wherein, the sampling frequency of the environment verification module for capturing wireless signal fingerprints is not less than 10Hz.

7. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, The distributed ledger establishes a data channel for internship tasks through a consensus mechanism; the distributed ledger associates and stores each query instruction initiated by the logical query module with the judgment result of the credit evaluation module to achieve the traceability of credit data.

8. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, The business monitoring module includes a silent data acquisition unit; the silent data acquisition unit is used to monitor the document processing node information of the tutor in the office collaboration system without collecting geographical location information, and convert it into business feature information.

9. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, After determining that the internship behavior is triggered by a genuine business transaction, the credit evaluation module triggers the distributed ledger to perform a credit record update operation. The credit record update operation includes binding the feature summary of the response voucher with the corresponding business evidence identifier on the blockchain and updating the accumulated credit score of the internship credit file.

10. The internship and employment information management system based on blockchain technology according to claim 1, characterized in that, The challenge instruction generated by the logic challenge module includes a verification token; the verification token is used to require the student terminal to simultaneously upload a snapshot of the real-time signal characteristics generated by the environment verification module when responding, as the input basis for determining the degree of overlap.

Images