Search

Taking Uyghur character recognition as an example, this paper use a method of image segmentation which combines traditional methods with CNN, and transforms and implements the Uyghur character on intelligent devices. Firstly, after analyzing the application of several general segmentation methods, this paper finds some shortcomings but also some ideas for Uighur segmentation. Then, starting from the characteristics of Uyghur language, such as structure, word formation and input habits, the author studies the idea of Uyghur adhesive language segmentation from the perspective of language characteristics, puts forward the basic algorithm of Uyghur symbol segmentation, and applies the Uyghur character adhesion segmentation based on minimum spanning tree and multi-queue primitive merging model to improve the segmentation efficiency. In addition, in order to solve the limitations of the traditional handwriting recognition framework of “preprocessing + feature extraction + classifier”, this paper puts forward a new solution of Uyghur handwriting recognition technology combining prior domain knowledge with CNN, constructs a Uyghur handwriting recognition model on the basis of CNN and random elastic deformation, and also improves the recognition rate of the system by combining domain knowledge with CNN. According to the experimental results, we can conclude that the mixed method proposed in this paper can effectively break the technical bottleneck of traditional methods, thus improving the efficiency of segmentation and recognition.

As the current society is increasingly facing major challenges from extremism and terrorism, protecting key urban public facilities and important targets from destruction is an important challenge facing the security departments of all countries. Based on real scene, this paper conducts researches on anti-terrorism security game algorithms and emergency response models in response to the three key links of before, during and after terrorist attacks. First of all, this paper constructs a multi-round joint attack game and emergency response model based on cooperation, establishes the optimization problem of solving the defender’s optimal strategy in mathematical form, and then obtains the optimal defense strategy. Secondly, in response to the fact that terrorists are not completely rational, a new hybrid model is constructed to propose an efficient allocation and scheduling algorithm for safe resources in response to terrorist attacks. Thirdly, a model of crowd evacuation strategy after a terrorist attack is built based on the problem of crowd evacuation in multiple rounds of premeditated cooperative attacks. Finally, taking the area of the first ring of a certain city as a real scene, a complete game system of the whole process is constructed, and the game effectiveness evaluation of the existing security resource allocation scheme in the first ring area is carried out. Through the research of this thesis, the author puts forward some new technical ideas for the current society’s anti-terrorism governance, and hopes to provide some technical references for the decision-making of security agencies.

With ever increasing complexity and intelligence of the modern cities, protecting key public facilities and important targets from any damage is a major challenge for the security sector. In all types of anti-terrorism prediction models, the prediction of attack behaviour is indispensable. Therefore, the attack behaviour model plays an important role in the anti-terrorism security system. This paper builds the attacker’s behaviour model, and carries out the prediction about the possible attack behaviour by the attacker model based on random strategy. According to the attack strategies, analysis and construction of the attack tree and attack graph are carried out based on the state-based stochastic model. The paper describes the security system in detail taking use of the state-based stochastic model method, so as to clarify the state distribution and the transfer relationship between the states of various security resources after threatened by attacks. At the same time, this paper applies the state-based stochastic model to establish the attacker model through the impact of attack on the security system.

Most existing recommender methods have insufficient power to capture or recover fine-grained features of products and user preferences. Knowledge graphs contain considerable information about products and mutual relationships in the world. To analyze the fine-grained features during the process of recommendation, we propose a knowledge-aware collaborative learning framework (KACL), which is the first work to combine users’ historical reviews with knowledge graphs. First, we use a named-entity recognition (NER) system to recognize the named entities that correspond to item features in the unstructured reviews. After that, we use an entity-linking (EL) system to map entities which identified in the first step to the corresponding entity in Wikipedia. Next, we constructed a sub-graph that depends on the extracted entity and the related ones and embedded the sub-graph into a low latent vector space as pretrain item embedding through the knowledge graph embedding model TransR. In the next step, we will fine-tune item embedding through deep learning built by 4 levels of relu and combine collaborative information with pretrain item embedding. And in accordance with the user review behavior, the item embedding will be averagely poled to express the user embedding, we will put the item embedding and the user embedding into the classification model. Finally, we use the KACL to integrate collaborative filtering with entity representation and make recommendations. Experiments are performed on real-world datasets to evaluate the effectiveness of our new method. The results show that the analysis of fine-grained features based on a knowledge graph helps our KACL improve recommender accuracy compared with the selected state-of-the-art models.

Finding similarity degree is one of the significant technologies used in the sample-based machine translation. It works in the following principle, first matching the input sentences with a sentence in the sample database, after that it is necessary to pick up parts of the similar sentences for the sentence which is aimed to translate; it is finished by correcting the structure or paraphrasing it with a relevant meaning. For that reason, the degree of similarity of two samples highly affects on the results of translation. Thus, there are dependence between quality of the outputs and the similarity degree.

This paper introduced the basic theory of the Support Vector Machine (SVM) and k-Nearest Neighbor (kNN) algorithm and two different features selection methods in Kazak natural language.An empirical study of using the SVM,kNN,Bayes algorithm to categorize the Kazak text was conducted.The experimental results show that compared with kNN,Bayes,SVM has better categorization of the Kazak text.Due to the characteristics of Kazak's morpheme and configuration,the precision and recall will be lowered if the word is cut with affix.