With the rapid expansion of digital asset trading, the contradiction between data sharing and privacy protection has increasingly become a significant challenge in the Internet environment. To address this issue, this paper proposes a secure multi-party computation scheme based on blockchain technology. Firstly, in response to the risk of data leakage in distributed storage scenarios, a threshold-based encryption algorithm is designed, utilizing a distributed key protection mechanism to effectively prevent single-point failures and data breaches. Secondly, a smart contract system is developed: the ERC721 contract is used to confirm the ownership of data assets, the ERC20 contract facilitates the transfer of usage rights, and the threshold decryption contract ensures secure multi-party computation and compliant incentive distribution. The collaboration of these three types of contracts enables comprehensive on-chain management of data assets, covering the entire process from ownership confirmation and circulation to compliant usage. In addition, this paper integrates non-interactive zero-knowledge proofs into the multi-party interaction process, allowing public verification of data consistency and computational validity on the blockchain. Finally, experiments are conducted to evaluate the impact of computation latency, communication overhead, and encryption parameters on system performance. The proposed scheme demonstrates significant performance improvements over mainstream SMPC protocols, with a 95.4% reduction in key generation time and a 19.5% reduction in ciphertext decryption time. Meanwhile, the scheme effectively resists various semi-malicious attacks, ensuring data security and privacy.

Federated learning in heterogeneous data scenarios faces two key challenges. First, the conflict between global models and local personalization complicates knowledge transfer and leads to feature misalignment, hindering effective personalization for clients. Second, the lack of dynamic adaptation in standard federated learning makes it difficult to handle highly heterogeneous and changing client data, reducing the global model’s generalization ability. To address these issues, this paper proposes pFedKA, a personalized federated learning framework integrating knowledge distillation and a dual-attention mechanism. On the client-side, a cross-attention module dynamically aligns global and local feature spaces using adaptive temperature coefficients to mitigate feature misalignment. On the server-side, a Gated Recurrent Unit-based attention network adaptively adjusts aggregation weights using cross-round historical states, providing more robust aggregation than static averaging in heterogeneous settings. Experimental results on CIFAR-10, CIFAR-100, and Shakespeare datasets demonstrate that pFedKA converges faster and with greater stability in heterogeneous scenarios. Furthermore, it significantly improves personalization accuracy compared to state-of-the-art personalized federated learning methods. Additionally, we demonstrate privacy guarantees by integrating pFedKA with DP-SGD, showing comparable privacy protection to FedAvg while maintaining high personalization accuracy.

The problem of data island hinders the application of big data in artificial intelligence model training, so researchers propose a federated learning framework. It enables model training without having to centralize all data in a central storage point. In the current horizontal federated learning scheme, each participant gets the final jointly trained model. No solution is proposed for scenarios where participants only provide training data in exchange for benefits, but do not care about the final jointly trained model. Therefore, this paper proposes a newboosted tree algorithm, calledRPBT (the originator Rights Protected federated Boosted Tree algorithm). Compared with the current horizontal federal learning algorithm, each participant will obtain the final jointly trained model. RPBT can guarantee that the local data of the participants will not be leaked, while the final jointly trained model cannot be obtained. It is worth mentioning that, from the perspective of the participants, the scheme uses the batch idea to make the participants participate in the training in random batches. Therefore, this scheme is more suitable for scenarios where a large number of participants are jointly modeling. Furthermore, a small number of participants will not actually participate in the joint training process. Therefore, the proposed scheme is more secure. Theoretical analysis and experimental evaluations show that RPBT is secure, accurate and efficient.

专业集群与产业链合作是工程教育回归工程实践的有效途径.为了充分发挥新一代信息技术专业集群的优势与特色,以提高信息类专业学生的工程实践与创新能力,提出基于专业集群的实践教学体系构建原则、实践教学体系、支撑条件和保障机制.实践表明,所提出的实践教学体系提高了学生在多学科专业背景下解决复杂工程问题的能力、综合实践能力、创新能力、团队协作能力和工业化素质.

At present, there are a large number of growing medical applications in the application market. It is difficult for users to find satisfactory medical services conveniently and efficiently. The classical collaborative filtering algorithm has some problems, such as cold start, unsatisfactory recommendation results, and so on. This paper proposes a hybrid medical service recommendation approach based on knowledge graph to solve the above problems. This approach introduces the open knowledge graph and establishes the semantic link relationship between the mobile application and the knowledge graph entity. It not only enhances the semantic feature of single application for improving the accuracy of recommendation results, but also realizes the in-depth analysis of the semantic relationship among multiple application entities in the knowledge graph through the TransHR model which can alleviate the cold start problem. Then we design a hybrid recommendation algorithm based on multi-dimensional similarity fusion. This algorithm uses the entropy method to organically integrate the calculation results of multi-dimensional semantic similarity, such as feature vector similarity, entity relation similarity, and user rating similarity. It is convenient and efficient to recommend satisfactory medical application services to target users. Finally, we test and analyze the accuracy and effectiveness of our proposed approach by experiment.

Anonymization technology is an important technology for privacy protection in the process of data release.Usually, before publishing data, the data publisher needs to use anonymization technology to anonymize the original data, and then publish the anonymized data.However, for data publishers who do not have or have less anonymized technical knowledge background, how to configure appropriate parameters for data with different characteristics has become a more difficult problem.In response to this problem, this paper adds a historical configuration scheme resource pool on the basis of the traditional anonymization process, and configuration parameters can be automatically recommended through the historical configuration scheme resource pool.On this basis, a privacy model hybrid recommendation algorithm for user satisfaction is formed.The algorithm includes a forward recommendation process and a reverse recommendation process, which can respectively perform data anonymization processing for users with different anonymization technical knowledge backgrounds.The privacy model hybrid recommendation algorithm for user satisfaction described in this paper is suitable for a wider population, providing a simpler, more efficient and automated solution for data anonymization, reducing data processing time and improving the quality of anonymized data, which enhances data protection capabilities.

In order to implement service selection efficiently，and to build a complex software service system that can meet the needs of users by using service aggregation method，this paper proposes a Hybrid Enhancement Artificial Bee Colony ( HEABC) algorithm. The algorithm combines K-means algorithm，K-Nearest Neighbor ( KNN) algorithm and ABC algorithm to ensure that ABC algorithm always maintains continuity when updating solutions in discrete solution space. The algorithm enhances the exploration and development capabilities of the bee colony by increasing the ability of information sharing between bee colonies. In terms of non-functional perception of software services，this paper introduces the concept of service contract to achieve more comprehensive user satisfaction and dynamic needs. The simulation experiment used 60 different sets of data and compared it to other algorithms in terms of quality and execution time. The results show that compared with other algorithms，this algorithm has improved the solution time and the solution quality.

人才培养质量持续改进机制是保障高等教育质量的基础,分析了当前关于人才培养质量持续改进机制的研究现状,讨论了人才培养质量持续改进机制的内涵,提出了"三闭环"人才培养质量持续改进机制模型,以哈尔滨理工大学软件工程专业为例介绍了构建多方参与的"三闭环"人才培养质量持续改进机制的具体实践.

新一代信息技术专业与传统的工科专业存在比较大的区别.针对信息技术专业多学科交叉融合以促进自身发展、用人单位急需复合型应用型工程技术人才以及学生个人发展的需要,提出专业集群建设的原则、新一代信息技术专业集群的构成关系和专业集群建设的保障措施,并介绍哈尔滨理工大学在专业集群建设方面取得的成效,最后指出在专业集群建设方面存在的困难和进一步的改进措施.

针对高校电子信息类专业中存在的人才培养与社会需求脱节、学生实践动手能力不足、教师缺乏工程实践经验和普遍存在的重学术轻实践以及企业难以参与人才培养全过程的问题,结合新工科的建设要求,建立了校企协同育人的新模式、具有专业特色的人才培养新方案、实践教学的新平台和新体系以及企业参与评价的校企共赢的长效合作机制,并制定了具体的人才培养实施方案,在集成电路设计与集成系统和电子科学与技术两个专业进行了实践探索,取得了良好的效果.

针对目前程序设计基础课程存在的教学方式单一化、考核环节集中化等问题,文章以OBE-CDIO教育理念为指导,对于该课程的课程教学模式、课程设计和课程考核方式等方面的改革进行了有益的探索与研究.我们将工程教育认证中学生的毕业要求作为制定教学目标的依据,将MOOC、雨课堂等现代化教学资源融入课堂教学,采用CDIO教学理念指导课程设计,实现教学环节与考核环节的科学化与多元化,全面地培养学生的综合实践能力.

文章针对软件工程专业现有课程体系中存在的人才培养与人才需求之间不能很好对接的问题,提出了面向解决复杂工程问题能力培养的课程体系建设思路,通过使用复杂软件工程问题"MES作业车间调度子系统"贯穿整个教学过程,重新进行了课程设计和教学设计,起到了较好的效果.

文章从"课程思政"的价值本源、学生自身发展的需要以及全程育人的理念出发,分析了在专业课程中实施"课程思政"的必要性,并以"软件项目管理"为例,探讨通过专业课修订教案大纲,添加思政元素,改革课程考核方式等措施,提高专业课课程思政的教学效果与质量.

地方高校在新工科应用型人才培养中凸显出教育理念陈旧、学科专业间存在壁垒、多方协同育人机制缺失、工程实践能力培养偏弱、人才培养缺少国际认同等问题.为了突破这些瓶颈,哈尔滨理工大学以满足国家与区域新兴产业人才需求为目标,在人才培养新模式、新体系、新机制和产学研用协同育人新平台等方面进行了初步探索,并取得了一定的成果.

针对深度神经网络在躲避攻击多目标对抗方法中输入的数据易导致机器误解码,提出一种深度神经网络结合蚁群算法的躲避攻击多目标对抗方法.设计一种与变换器和多个模型组成的体系结构,利用变换器生成一个多目标的对抗性样本,利用深度学习训练的分类器对输入值进行分类;引入蚁群算法,利用蚂蚁互相交流学习的正反馈原理保证算法的收敛性和寻优速度;融合两种算法的优势,实现躲避攻击的多目标对抗.实验结果表明,相比其他现有方法,该方法在躲避攻击多目标对抗方面更具优势,实现了100％的攻击成功率.

Aiming at how to efficiently handle the massive temporal sequential manufacturing service request in cloud manufacturing service platform, this paper proposes a service response timeoriented service request segmentation algorithm. The manufacturing task similarity distance algorithm is used to guarantee the segmentation algorithm accuracy. Then the correlated region is constructed under the premise of ensuring the integrity of manufacturing service request. On the basis of this, a manufacturing service resources allocation algorithm based on correlated region is proposed so as to ensure the equal distribution of manufacturing service resources in temporal dimensions, and then to improve the efficiency of the global optimization of the manufacturing service resources in the case of limited resources. Finally, the correctness and validity of the algorithm are proved by experiment and analysis.

In order to better solve the problem that the remaining life of cutting tool is difficult to predict accurately, this paper studies three aspects of the selection of monitoring indexes, the extraction of data features and the establishing of prediction models Firstly, Cutting force and vibration frequency were selected as the indirect monitoring indexes of cutting tool These two indexes can accurately reflect the state of cutting tool, and also can solve the problem that the selecting the direct monitoring indexes causes, the wear analysis results of cutting tool being too subjective in the traditional state monitoring method Secondly, feature extraction is carried out by using wavelet packet analysis, and then the entropy values of the monitoring data are obtained They are taken as the input data Thirdly, the input data are used as the training data and testing data of the prediction model based on Deep Neural Network (DNN) Finally, the simulation experiments of the prediction method are carried out by using the real data of the workshop The results show that the model can effectively predict the useful life

In order to ensure the reliable work of electronic components in the marine cabinet, electronic equipment should be placed in a closed cabinet to prevent damage by sea dust, corrosive gases and rainwater. In this paper, based on the working principle of embedded miniature heat pipe, a heat transfer model of embedded miniature heat pipe based on irregular cross section is proposed to replace the traditional heat transfer device and solve the heat transfer problem in the marine cabinet. The heat pipe structure is analyzed and the heat capacity and thermal resistance network method is used to solve the thermal problem of the marine cabinet. Based on the analysis of heat transfer network of marine cabinets, the heat transfer model of embedded miniature heat pipes is established by dividing the nodes of heat pipes. According to the number of axial nodes of the heat pipe, the heat transfer power of the heat pipe is solved. The experimental results show that the heat resistance of the miniature heat pipe embedded in the marine cabinet can be divided into the cooling thermal resistance and the cyclic thermal resistance, in which the proportion of the cooling area and the length of the cooling section to the total length is the main factor affecting the cooling thermal resistance. Heat source temperature and liquid filling rate are the main factors affecting the thermal resistance. The cooling temperature has little effect on the heat transfer resistance. The heat transfer power can be improved by increasing the cooling area, length ratio of cooling section, cooling flow rate, heat source temperature and filling rate as well as reducing the cooling temperature. The first 3 factors are the best way to improve the heat transfer power.

In the cloud manufacturing environment, the flexible job shop scheduling needs to handle the online task and offline tasks simultaneously, which make the processing time of equipment become from continuous to discrete. For solving this issue, a mathematical model considering discrete equipment capability was proposed. In this model, the additional profits constraint and the equipment capability conflict constraint are introduced. The former was used to promise the optimal solution is profitable for the manufacturing enterprise; the latter was used to promise the online task coming from the other manufacturing enterprises can be processed. Then an improved ant colony algorithm is adopted to solve this problem. Finally, the experimental results show that the flexible job shop scheduling problem with discrete equipment capability in cloud manufacturing environment could be solved effectively.