A personalized exercise recommendation method that combines cognitive diagnosis and deep factorization machine was proposed to address the problems of single modeling angle and unreasonable exercise recommendation results of the existing exercise recommendation based on cognitive diagnosis. Firstly， a new method for calculating the relationship between knowledge points was designed to construct a course knowledge tree， and the concept of enhanced Q matrix to accurately represent the relationship between knowledge points contained in exercises was proposed. Secondly， the Neural Cognitive Diagnosis with Knowledge-based Discernment （NeuralCD-KD） model was proposed to calculate the enhanced Q matrix. In the model， the feature second-order cross and attention mechanism were used to fuse internal and external factors of exercise difficulty， and the students’ cognitive states were simulated. The effectiveness of the proposed cognitive diagnosis model was verified on private and public datasets， and this method was able to give reasonable explanations for students’ cognitive states. To personalize exercise recommendation， a Neural Knowledge-based Cognitive Diagnosis with Deep Bilinear Factorization Machine （NKD-DBFM） method was proposed by combining the diagnostic model with deep bilinear factorization machine， and the effectiveness of this proposed exercise recommendation method was verified on the private dataset. Compared with the optimal baseline model Neural Cognitive Diagnosis Model （NeuralCDM）， the proposed method improves the Area Under Curve （AUC） by 3.7 percentage points.

In the current multi-level secret sharing scheme, the presence of higher-level participants could be replaced by several lower-level participants. As to the multi-secret sharing scheme, secrets could only be shared under the same level. To eliminate the two limitations, based on Birkhoff Interpolation and the intractability of discrete logarithm, a new verifiable hierarchical threshold multi-secret sharing scheme was proposed. This scheme simultaneously maintained multiple levels and multiple secrets for every threshold level. Every participant kept one sub-secret only, which was convenient to manage and use.

A multi-secret threshold secret sharing scheme was introduced. Through one process of secret sharing, any secret sharing could be achieved, and participants share a secret and the length is only a secret length. Meanwhile, considering such programs threshold of safety, a secure threshold multi-secret sharing scheme based on special difference equation was given. Analysis shows that the given threshold secret sharing scheme has the information rate of 1/2, and for fraud prevention is unconditionally safe.

A dynamic and verifiable threshold color visual multi-secret sharing scheme was proposed. The scheme shared many secret images so that any t or more sub-secrets could recover the secret image, but any t-1 or fewer of them gained no information about the original secret image. In the scheme, multiple secrets can be shared in one sharing session. The sub-secret of each participant can be used by many times，and the secrets can be renewed without renewing the sub-secrets of the participants. What is more, a participant can be flexibly accepted or fired.

Key management is an important field in information security. Key agreement is one of the core problems in the key management．A new computationally secure threshold multi-secret key agreement scheme with authority based on the theory of linear codes was introduced．The scheme makes the users share the secret key, and does not need to be based on discrete logarithm. The set of participants to perform key agreement should satisfy threshold. In the scheme, multi-secret key can be initiated in one key agreement session. The scheme based on linear codes has the property of authority by checking the equality to prevent the third party attack.