Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-04-01 Cooperative journals: 《计算机应用研究》
Abstract： Traditional content-based recommendation algorithms have lower accuracy, while data sparseness and cold start problems are common in collaborative filtering recommendation algorithms. To solve this problem, this paper proposed a hybrid recommendation algorithm based on content and collaborative matrix factorization technique. The algorithm realized the decomposition of content and collaborative matrix in a common low-dimensional space while preserving the local data structure. This paper used an iterative method based on multiplication update rules in parameter optimization, improved learning ability. The experimental results show that the proposed algorithm is superior to other representative projects cold start recommendation algorithm, which effectively alleviates the data sparseness and improves the efficiency of the algorithm.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-01-28 Cooperative journals: 《计算机应用研究》
Abstract： For the problem of the singular value decomposition (SVD) watermarking algorithm caused high false alarm rate and low robustness, this paper proposed an adaptive zero watermarking algorithm , which was based on block NMF and boost normed singular value decomposition. First, the original gray image was transformed into two levels of DWT transform, then after the transformation of the LL2 subbands were divided into non-overlapping blocks, and each sub-block was broken down into NMF with rank r, then boost normed singular value decomposition was used for eigenmatrix derived from NMF decomposition. According to the relationship between the maximum singular value of each block matrix and the mean value of the global maximum singular value, it constructed feature-vector . Finally the generated feature vector was xor operation with the random encrypted watermark image of Arnold and the logistic map to generate a zero watermark. The parameter β in the singular value matrix must be determined adaptively by the optimization algorithm (BAS) , and find out the scaling ratio of the most resistant attack. The experimental results show that, in the false alarm problem, the NC value is below 0.4. Under the condition of JPEG compression, noise, filtering, rotation, shearing and mixed attack, the normalized coefficient NC of the extracted watermark image and the original watermark image can reach more than 99%. The method can solve the false positive errors efficiently and has strong robustness. Can resist all kinds of attacks effectively.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-01-28 Cooperative journals: 《计算机应用研究》
Abstract： To overcome the local optimum and premature convergence due to loss of population diversity of the particle swarm optimization algorithm, this paper proposed a disruption particle swarm optimization algorithm based on exponential decay weight (EDW-DPSO) . Firstly, the population was semi-uniformly initialized to distribute the population in an overall uniform, locally random manner. Secondly, the dynamic splitting operator was introduced to perform splitting operations on particles which satisfying the splitting condition, increasing the diversity of the population and avoiding the particles falling into local optimum. Finally, the exponential decreasing inertia weight was used to balance the global search and local development ability of the particles. The experimental results show that the algorithm has a large search space in the early stage, and the population diversity increases. In the later stage, emphasizeing the local development to improve the convergence precision and optimization ability. It can also accelerate particles jumping out of the local extremum and approximate globle optimum.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-01-28 Cooperative journals: 《计算机应用研究》
Abstract： Present methods of opinion target extraction fail to extract multi-targets and compound target based on supervised learning model. This paper proposes an opinion method to deal with this issue based on syntactic structure and dependency relation. Firstly, it analysed the dependency relationships between opinion targets and opinion words based on the different syntactic constituents of opinion targets and opinion words. Then, it defined features according to different dependencies. Finally, it selected the optimal feature combination by the greedy feature selection method to extract opinion targets with the conditional random field model. It conducted experiments in the evaluation data of task 3 of COAE2011. The results show that the value of F1 is 3%-6% higher than that of the present method of opinion target extraction and the method can identify the opinion targets effectively.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-10-11 Cooperative journals: 《计算机应用研究》
Abstract： Association rules are attracting wide attention in the field of big data mining. The key and difficult point of the algorithm is to mine frequent sets. In order to further improve the speed of the association rules mining frequent sets and optimize the execution performance of the algorithm, an association rule mining algorithm based on improved memory structure is proposed. For the existing algorithm, the storage structure is simple, the candidate set with a large amount of redundancy is generated, the time and space complexity is high, and the mining efficiency is not ideal. The algorithm of this paper is based on the Spark distributed framework. The partitions are mined in parallel to extract frequent sets. It is proposed to use the Bloom filter to store the project in the mining process, and to simplify the operation of the transaction set and the candidate set, so as to optimize the speed of mining frequent sets. Save computing resources. Compared with the YAFIM algorithm and the MRApriori algorithm, the algorithm has a significant improvement in the efficiency of mining frequent sets under the condition of occupying less memory. The algorithm can not only improve the mining speed, reduce the memory pressure, but also has good scalability, so that the algorithm can be applied to larger data sets and clusters, so as to optimize the performance of the algorithm.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-10-11 Cooperative journals: 《计算机应用研究》
Abstract： Aiming at the inaccuracy of embedding watermarking location and random selection of embedding intensity parameters, this paper proposed a nonlinear scale space adaptive equalization watermarking algorithm. It used KAZE algorithm to extract and filter the feature points with strong stability in nonlinear scale space for constructing the watermark region. Then it used singula value decomposition on watermarking image, constructed a new matrix as the watermarking carrier to be embedded and calculating the embedding intensity by adjusting the fitness function of the ftuit fly optimization algorithm, and with the DWT-SVD algorithm, the watermark embedding process is adaptively completed. Extracting the feature points from the under attack watermark image to synthesize the feature region matrix. And it used inverse process to extract watermark image. The experimental results show that the PSNR values are above 44dB and the average NC value is as high as 0.99, which effectively equalizes the invisibility and robustness of the watermarking algorithm.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-10-11 Cooperative journals: 《计算机应用研究》
Abstract： Hyperchaotic system has certain advantages because of its characteristics of sercret key space and sensitivity to initial value. Aiming at the low computation efficiency and poor security of the existing Zero watermarking algorithm, this paper proposed a zero watermark algorithm based on hyperchaotic system. Firstly the algorthm realizes the encryption pretreatment of watermarking information taking advantage of the large key space and sensitivity of initial value of the Chen hyper-chaos system. And then analyzing the influence of each bit plane on the image, least significant bit of the image carrier is initialized to zeros to extract the feature matrix using of block mean binary quantization method. Finally, Arnold scrambled characteristics matrix and hyper-chaotic encrypted binary watermarking are executed for xor operation to construct a zero watermarking. The simulation attack experiment and comparison with the precious zero watermark algorithm show that the algorithm in this paper can resist noise attack, filter attack, compression attack and shear attack while maintaining better robustness.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-07-09 Cooperative journals: 《计算机应用研究》
Abstract： In order to solve the problems such as sparse data, cold start and lack of diversity of recommendation results in traditional collaborative filtering recommendation algorithms, this paper proposes a collaborative filtering recommendation algorithm that integrates social networking with key users. Based on the score matrix of user projects, the algorithm integrates user social networks to derive social trust matrix, integrates key user information to obtain key user scoring matrix, and then uses these three matrix data distributions to predict user's target project with different weights. score. At the same time, aiming at the massive data problem, this paper uses the Spark distributed cluster to realize the parallelization of the algorithm. The experimental results show that the algorithm can effectively alleviate the data sparse problem and improve the data processing speed and recommendation accuracy.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-06-19 Cooperative journals: 《计算机应用研究》
Abstract： To solve the problems of the zero-watermarking algorithms with weak against geometric attack, this paper proposed a zero-watermarking algorithm against geometric rotation attack. Firstly, the algorithm determined the near-destructive maximum inscribed square region with approximately pixels lossless in the center area, according to the pixel distortion of the image of the scale-invariant feature transform (SIFT) rotation correction. Then, the square region performed two-level redundant discrete wavelet transformation and extracted the low-frequency region. What’s more, it extracted the largest singular value of each block to construct a transition matrix into block processing from the low-frequency area. Next, to obtain a characteristic matrix, it compared the value of each element in the transition matrix with its average value. Finally, it used the watermark image and constructed the characteristic matrix a zero watermark. The experimental results show that compared with the only rotation correction algorithm of SIFT, the robustness against rotation attack is up by an average of 13.26%. Compare with the rotation corrections algorithms of the GH rotation moment and pseudo-Zernike orthogonal moment, the anti-rotation attack robustness is up by 1.1% and 0.94% respectively. Also, it has a good effect on common conventional attacks, scaling attacks, cyclic translation and small-scale shear attacks.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-20 Cooperative journals: 《计算机应用研究》
Abstract： Aiming at the problem of hyperspectral image collaborative representation anomaly detection algorithm that the output of the central pixel is small and difficult to distinguish from the background when dual window center pixel is anomalous pixel and background dictionary contains the same kind of anomalous pixels. This paper proposed an improved collaborative representation for hyperspectral imagery anomaly detection algorithm. In order to reduce the weights of the anomalous pixels in the background dictionary, using the distance between the atom and the mean of the background dictionary to adjust the weights of the atoms, so as to increase the output of the central pixel in the above conditions. Experimental results show that the proposed algorithm achieves better detection results with different dual windows, and verifies the effectiveness of the proposed algorithm.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-18 Cooperative journals: 《计算机应用研究》
Abstract： The certificateless signcryption scheme effectively solved the key escrow problem in identity based signcryption scheme while kept its certificate-free property. Aiming at the low computation efficiency and poor security of the existing certificateless signcryption scheme, this paper proposed a new certificateless signcryption scheme without pairings based on a sort of secure signature scheme. The scheme used binding the hash functions with identities of users and the method of combing the public and private key to generate a new key. The scheme was confidential and unforgeable based on the hard problem of discrete爈ogarithm on the elliptic curve under the random oracle model. Compared爓ith爀xisting爏chemes, the proposed爉ethod爄mproves爐he爀fficiency without爑sing燽ilinear爌airing燼nd爀xponential爋peration爑nder爐he爏ecure爏ituation.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-18 Cooperative journals: 《计算机应用研究》
Abstract： In order to better perform the characteristics of natural image curves and further improve the robustness of digital watermarking algorithms, this paper proposed a robust zero-watermarking algorithm based on Curvelet-DSVD and visual cryptography. Firstly, performing Arnold scrambling on the original image, then obtaining the low frequency domain information by Curvelet transform, segmenting the low frequency domain information, and each block was double singular value decomposition(DSVD) , by using the relationship between the maximum singular value of the block and the mean value of the global singular value structuring characteristic matrix. At the same time, this paper used the visual cryptography to generate two shared copies of the watermark information. Finally, one of the shares was execute to Arnold scramble and then xor operation with the feature matrix to generate a zero-watermark. Experimental results show that the proposed algorithm can effectively resist the conventional attacks and is more robust and more secure than the existing zero-watermarking algorithms.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-18 Cooperative journals: 《计算机应用研究》
Abstract： The existing subspace clustering algorithms cannot balance the density of the data in the same subspace and the sparsity of the data between different subspaces and most algorithms cannot solve the overlap of data. To solve the above problems, this paper proposed a novel algorithm of overlapping subspace clustering algorithm under sparse condition (OSCSC) . The algorithm used the mixed norm representation method of L1 norm and Frobenius norm to establish the subspace representation model, and the weighted L1 norm regular term could improve the sparsity of different subspaces and the density of the same subspace. Then, the algorithm performed rechecks on the partitioned subspaces by using an overlapping probability model subject to exponential family distribution to determine whether exist overlapping in different subspaces, which could further improve the accuracy of clustering. The results of the experiment on both artificial datasets and real-world datasets show that the algorithm has better clustering performance by being compared to other contrast algorithms.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-02 Cooperative journals: 《计算机应用研究》
Abstract： The assessment of urban accessibility has always been a hot topic of concern in the research field of smart transportation. The traditional regional accessibility assessment model generally only supports the single dimension data of GIS or GPS as the basic data for the assessment of accessibility, so it is impossible to avoid the problem of inaccurate assessment of regional accessibility due to the influence of external factors. Aiming at this problem, this paper constructs a city area accessibility evaluation model to support multivariate data using the multidimensional data such as GPS vehicle traffic data, time and weather as the basis of regional accessibility. On this basis, the calculation model of the region accessibility ratio based on multidimensional OD matrix is designed in this work which is used as the quantitative methods of regional accessibility to achieve the purpose of improving the accuracy of accessibility assessment. In addition, to solve the problem of traditional GPS data cleaning method, such as effective information missing and inaccurate data correction, which is caused by its over-roughness, the serial data cleaning method based on the statistical theory is applied in this model. The speed and acceleration information of the Taxi GPS data is considered in this data cleaning method to correct the potential error and to improve the GPS data cleaning effect. Experiment result shows that the accuracy of the regional accessibility calculated by using the multivariate data urban area accessibility assessment model proposed in this paper is 9.1% -37.8 higher than that of the traditional methods, and the accuracy of the regional accessibility assessment ratio and travel time are increased by 12.6% -35.5% and 18.5% -31.6% respectively.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-19 Cooperative journals: 《计算机应用研究》
Abstract： This paper proposed a hybrid flower pollination algorithm (refrHFPA) based on refraction principle for the slow convergence rate of flower pollination algorithm and low optimization accuracy. The algorithm firstly used the harmony search algorithm to improve the convergence speed of the algorithm, then it used the refraction principle to improve the diversity of the population, and helped the algorithm to jump out of the local optimal and improved the accuracy of optimization. It used eight test functions to compare other intelligent algorithms, and the results show that the refrHFPA algorithm has a significant improvement in convergence speed and optimization accuracy.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-17 Cooperative journals: 《计算机应用研究》
Abstract： Area selection based on the digital watermarking algorithm is not enough to reflect the image of important information, lead to decreased robustness problems, this paper put forward a measure space characteristics of the area the strong robustness of watermarking algorithm. Through scale space feature point detection and extraction characteristics near the image center of gravity and non-overlapping regions, regional matrix synthesis characteristics, with the transform domain watermarking algorithm embedding watermarking, the area may contain the watermark image feature extraction under attack, the characteristics of regional matrix synthesis, it used inverse process to extract the watermark embedding watermarking algorithm. Experiments show that this algorithm is not only robust to a series of attacks, but also has good invisibility.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-12 Cooperative journals: 《计算机应用研究》
Abstract： Concerning the problems of high time complexity, large storage space requirements and low accuracy when traditional clustering algorithm cluster stream data, this paper proposed a kind of stream data clustering algorithm based on differential sampling. First, it used the differential sampling method sampled stream data, and used sample points to construct kernel matrix. Then it used kernel fuzzy C-means clustering algorithm clustered the data points in the kernel matrix, obtained a marked sample kernel matrix. Finally, using the marked kernel matrix divided the stream data. Meanwhile, this paper adopted the fading cluster mechanism to update kernel matrix in real time. Experimental results show that compared with the traditional clustering algorithm, the proposed algorithm achieves lower time complexity, real-time clustering at the same time, get the ideal clustering result.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-12 Cooperative journals: 《计算机应用研究》
Abstract： The Word2Vec model is a neural network model (NNLM) that converts words in text into a word vector. It is widely used in natural language processing tasks such as emotional analysis, question answering robot and so on. Word vectors generated for the Word2Vec model lacked the ambiguity of context and the inability to create OOV word vectors. Based on the similarity information of document context and Word2Vec model, this paper proposed a word vector generation model that conforms to the meaning of OOV context. It is called the Word2Vec-ACV model. The model was similar to the process of the word vector generated by the Word2Vec model, but it was different. First of all, Word2Vec model of the continuous word bag (CBOW) and the Hierarchical Softmax trained the word vector matrix, namely the weight matrix. Secondly, the co-occurrence matrix was normalized to get the average context word vector. Then, the word vector consisted of an average context word vector matrix. Finally, the vector matrix of the average context word vector matrix and the weight matrix were multiplied to get the word vector matrix. In order to simultaneously solved the ambiguity problem of out of vocabulary words and out of vocabulary words to create. In this paper, the average context word vectors were divided into two kinds: the global average context word vector (global ACV) and the local average context word vector (local ACV) . In addition, the two taken the weight value to form a new average context word vector matrix. The Word2Vec model can effectively express the word in vector form. Experiments on analogical tasks and named entity recognition (NER) tasks respectively, the results show that the Word2Vec-ACV model is superior to the Word2Vec model in the accurate expression of the word vector. It is a word vector representation method to create a contextual context for OOV words.
Hits
Hits
Downloads
Comment