Search

A frame synchronization algorithm based on delay-Doppler domain was proposed for orthogonal time frequency and space (OTFS) modulation systems with a single pilot structure. The signal sparsity and the relative position change of the pilot in the delay-Doppler domain of the received OTFS signals were utilized to identify the frame synchronization location. Simulation results were presented to demonstrate that the proposed scheme outperformed the classic synchronization scheme based on the preamble, which was commonly used in orthogonal frequency division multiplexing (OFDM) systems, in terms of lower synchronization failure rate under high signal-to-noise ratio (SNR) regions. Additionally, the proposed scheme exhibited greater performance advantages over the preamble-based scheme as the moving speed increases, making it a more suitable option for high-speed or even ultra-high-speed mobile communication scenarios.

Forward-looking imaging is crucial in many civil and military fields, such as precision guidance, autonomous landing, and autonomous driving. The forward-looking imaging performance of airborne radar may deteriorate significantly due to the constraint of the Doppler history. The deconvolution method can be used to improve the quality of forward-looking imaging; however, it will not work well for complex imaging scenes. To solve the problem of scene sparsity measurement and characterization in complex forward-looking imaging configurations, an efficient probability model-driven airborne Bayesian forward-looking super-resolution imaging algorithm is proposed for multitarget scenarios to improve the azimuth resolution. First, the data dimension of the forward-looking imaging scene was expanded from single-frame to multiframe spaces to enhance the sparsity of the imaging scene. Then, the sparse characteristics of the imaging scene were statistically modeled using the generalized Gaussian probability model. Finally, the super-resolution imaging problem was solved using the Bayesian framework. Because the sparsity characterization parameters are embedded in the entire process of imaging, the forward-looking imaging parameters will be updated during each iteration. The effectiveness of the proposed algorithm was verified using simulation and real data.

Aiming at the problem that current feature selection methods were still affected by noise and cannot effectively unify clustering and reconstruction effects, a robust feature selection method was proposed.A robust reconstruction error term was built by making the difference between low-rank reconstruction and projection reconstruction.After that, the features for clustering were selected from the reconstructed data instead of the original data.The learning of clean data and feature selection technique are allowed for joint learning and promote each other, thereby improving the robustness of the method on noisy data, and effectively unifying reconstruction and clustering.Compared with several kinds of graph embedding feature selection and reconstruction feature selection methods on five datasets, the experimental results showed that, except for the LUNG noise dataset, the proposed method outperforms the comparative feature selection method under both evaluation indicators (ACC and NMI).

Mining community structure in attribute network is helpful to further analysis of network nodes, which has important practical significance. Graph convolution neural network can effectively embed the structural information of attribute network and obtain the feature representation of nodes. Based on this, the community structure with good performance can be obtained. However, most of the existing graph convolution methods use fixed low-order graph convolution, only consider the first-order or second-order neighbors of each node, don’t make full use of the node relationship and ignore the diversity of network structure. In addition, the sparsity of the original network structure can’t be overcome, which will reduce the performance of community detection. In order to solve the above problems, this paper proposes a K-order graph convolution community detection method (KGCN) which combines attribute information and structure information. This method can effectively overcome the sparsity of the original network and use the high-order structure of nodes for community detection. Firstly, the original network is reconstructed according to the attribute information of nodes to alleviate the sparsity of the original network. Secondly, considering the high-order structure correlation, the K-order graph convolution encoder is used to encode the nodes to obtain the feature representation of the nodes. Finally, community detection is carried out using spectral clustering algorithm. Experimental results show that the KGCN method achieves better community detection results than the existing algorithms on four real datasets.

With the enrichment of data sources,data can be obtained easily but with low quality,resulting that the MVs are ubi-quitous and hard to avoid.Consequently,MV imputation has become one of the classical problems in the field of data quality mana-gement.However,most existing MV imputation approaches are proposed for static data,which cannot handle dynamic data streams arriving at high-speed.Moreover,they do not consider data sparsity and heterogeneity simultaneously.Therefore,a novel MV imputation approach,real-time imputation based on individual models (RIIM) is proposed.In RIIM,the MVs are effectively filled by combining the basic ideas of neighbors-based imputation and regression-based imputation with consideration of sparsity and heterogeneity of data.For the dynamic and real time of data streams,the MV imputation model is updated incrementally.Moreover,an adaptive and periodic updating strategy for optimal neighbors search is proposed to solve the problem of high time cost and hard to determine the number of neighbors.Finally,the effectiveness of the proposed RIIM is evaluated based on extensive experiments over real-world datasets.

Copyright of Journal on Communication / Tongxin Xuebao is the property of Journal on Communications Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Collaborative filtering(CF) recommendation algorithm is widely used because of its reasonable interpretability and simple process.However,datasets in recommendation systems have the characteristics of large scale,high sparsity and high dimensionality,which bring a great challenge for CF recommendation algorithms.To alleviate the above issues,this paper proposes a collaborative filtering recommendation algorithm based on the rating region subspace(RRS).According to the user-item rating matrix,RRS firstly divides the scoring range into three different regions:high scoring region,medium scoring region and low scoring region.On the basis of these three regions,each user finds its item subspaces,that is,high rating subspace,medium rating subspace and low rating subspace.A new similarity measurement method is defined to calculate the rating support between users in each region subspace.Only if the rating supports of users in all subspaces are high,the users are similar,which avoids the ra-ting interference of lazy users.Experimental results show that the proposed method can solve the issue of data sparsity to a certain extent,reduce the computational complexity and improve the recommendation performance,especially on high-dimensional datasets.

Copyright of Journal of South China University of Technology (Natural Science Edition) is the property of South China University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Signal Processing is the property of Journal of Signal Processing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Frontiers of Computer Science & Technology is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Test & Measurement Technology is the property of Publishing Center of North University of China and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The imputation-based solution can alleviate the sparsity problem of recommendation system. Improved Jaccard similarity based on logistic user habit rating weighting, which is called the HabJac. By combining this metric, the K nearest neighbor ( KNN) obtains the imputation value. Furthermore, this paper proposed a new HISVD recommendation algorithm by combining the regularized singular value decomposition ( RSVD) technology to predict the unknown ratings. Firstly, the user' s habit rating was the most frequent one. Secondly, logistic weight was positively correlated with Euclidean distance between rating and habit one . The experimental results on four real data sets show that: a ) the optimal parameters of HISVD algorithm for different data sets are similar; b) HISVD surpasses the mainstream competitors for the MAE and RSME. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban) is the property of Zhongguo Xue shu qi Kan (Guang Pan Ban) Dian zi Za zhi She and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Information & Control is the property of Gai Kan Bian Wei Hui and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Xinyang Normal University Natural Science Edition is the property of Journal of Xinyang Normal University Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

A convolutional neural network(CNN) inference system is designed based on the FPGA platform for the problem that the convolutional neural network infers at low speed and it is power consuming on the general CPU and GPU platforms. By computing resource reusing, parallel processing of data and pipeline design, it greatly improved the computing speed, and reduced the use of computing and storage resources by model compression and sparse matrix multipliers using the sparseness of the fully connected layer. The system uses the ORL face database. The experimental results show that the model inference performance is 10.24 times of the CPU, 3.08 times of the GPU and 1.56 times of the benchmark version at the working frequency of 100 MHz, and the power is less than 2 W. When the model is compressed by 4 times, the system identification accuracy is 95%.

Aiming at the sparsity and cold start of user item score matrix in the recommendation system, a new algorithm is proposed by using the gray wolf algorithm (GWO) to optimize the support vector machine (SVM) classifier. Convolution matrix factorization model (ConvMF) is used to extract the effective feature vector of the user-item data, reduce the dimensionality, and form a sample set. Then the sample set is used for the classification and recognition of the vector machine, and the error points are automatically detected, and the gray wolf algorithm is updated iteratively to find the global optimal value, so that the corresponding products with different characteristics can be assigned according to user behavior. The study results show that the gray wolf algorithm optimized support vector machine is better than the traditional grid search algorithm (GS), particle swarm optimization algorithm (PSO) and other classification models. It can more accurately and effectively optimize the identification of recommendation results, reduce the error rate, and improve the prediction accuracy and recommendation accuracy. [ABSTRACT FROM AUTHOR]

The grid-based compressive sensing (CS) algorithm suffers from grid mismatch and is prone to artifacts in the case of insufficient resolution. The gridless CS algorithm is commonly used in two-dimensional harmonic estimation problems and is not suitable for complex signal models such as cross terms. In this regard, an alternating descent conditional gradient forward-looking imaging algorithm is proposed. The proposed algorithm firstly obtains a coarse estimate of the scattering point parameters with each iteration, and updates the parameter set. And then, the proposed algorithm performs gradient descent on the updated parameter set to obtain a fine estimate of the parameter set, realizing two-dimensional high-resolution imaging with continuous parameters under complex signal models. Simulation experiments illustrate the superiority and effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Copyright of Systems Engineering & Electronics is the property of Journal of Systems Engineering & Electronics Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Chongqing University of Posts & Telecommunications (Natural Science Edition) is the property of Chongqing University of Posts & Telecommunications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Concerning the issue that the convolutional neural network (CNN) accelerator design on most FPGA ends fails to effectively use the sparsity and considering both bandwidth and energy consumption,two improved CNN calculation optimization strategies based on linear systolic array architecture are proposed.Firstly,convolution is transformed into matrix multiplication to take advantage of sparsity.Secondly,in order to solve the problem of large I/O demand in traditional parallel matrix multiplier,linear systolic array is used to improve the design.Finally,a CNN acceleration comparative analysis of the advantages and disadvantages between parallel matrix multiplier and two improved linear systolic arrays is presented.Theoretical proof and analysis show that compared with the parallel matrix multiplier,the two improved linear systolic arrays make full use of sparsity,and have the advantages of less energy consumption and less I/O bandwidth occupation.

Feature selection is a common dimension reduction approach for processing high-dimensional big data, but it often involves multiple conflicting feature subsets evaluation objectives which are difficult to balance . To reach a compromise among various feature subset evaluations in feature selection and optimize the performance of subset, this paper proposed a subset evaluation multi-objective optimization based feature selection framework and focused on the application of multi-objective particle swarm optimization( MOPSO) in feature subset evaluation. The framework used sparsity, classification ability and information loss to design multi-objective optimization functions. Then it optimized the weight vectors of the features based on multi-objective optimization algorithm, and selected the " knee" of Pareto solution set as optimal vector. Finally, the framework realized feature selection based on weight vector ranking. This paper designed experiments to compare the performance of MOPSO based feature selection( FS_MOPSO) with four classical methods. The results on several standard data sets show that, FS_MOPSO shows higher classification accuracy in low dimensional space while ensuring less information loss. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban) is the property of Zhongguo Xue shu qi Kan (Guang Pan Ban) Dian zi Za zhi She and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Social recommendation based on matrix factorization enhances the learning accuracy by adding user trust relationship, but ignored the influence of related information between items on user interest in the process of model decomposition. This paper first provided an improved item similarity function to measure correlations between items by considering the user frequency factor, and then established a matrix factorization model with two regularization terms, which combined item relations regularization and social regularization into matrix factorization objective function to present the constraints on low rank approximation. It explored a major impact of the item association information on recommendation when optimizing the implicit feature matrix. The experiments on two real-world datasets demonstrate that this method outperforms other state-of-the-art matrix factorization approaches especially dealing with the large sparse data, and effectively alleviates the cold-start user problem. [ABSTRACT FROM AUTHOR]

Copyright of Information & Control is the property of Gai Kan Bian Wei Hui and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Advanced Engineering Science / Gongcheng Kexue Yu Jishu is the property of Advanced Engineering Science Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The computation of the classification process of Twin Support Vector Machines (TSVMs) is proportional to the number of samples. When the number of samples is large, the classification process will be time-consuming. In order to improve the sparsity of sample sets, a Fast Classification algorithm of Twin Support Vector Machines based on Affinity Propagation clustering (FCTSVMs-AP) is proposed. FCTSVMs-AP first performs adaptive AP clustering on the original data set. The center of the cluster is be used as the new sample set after reduction. According to the principle of minimum classification error, an optimization model is constructed. Moreover, the coefficients of the new decision tree is solved by quadratic programming. Furthermore, we prove that, when the sample set is compressed, the error between the new fast decision function and the original decision function is equivalent to the adaptive AP clustering of the original data set in the sample space. Experiments on artificial datasets and UCI datasets show that FCTSVMs-AP can improve the classification speed by effectively com pressing the number of samples, while the loss of classification accuracy is not statistically significant. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Signal Processing is the property of Journal of Signal Processing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Chongqing University of Posts & Telecommunications (Natural Science Edition) is the property of Chongqing University of Posts & Telecommunications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Telecommunication Engineering is the property of Telecommunication Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Aiming at optimizing and improving a hybrid collaborative filtering based on Spark platform for its sparsity, scalability and personalized recommendation by using the method of algorithm integration, this paper took the model of Stacking integration to integrate multiple weak recommender units in a linearly weighted into a comprehensive recommender. Firstly, this algorithm optimized the collaborative filtering based on the nearest neighbor by presorting and adjusting the similarity calculation strategy with popularity and praise degree, and improved the rationality and complexity of similarity calculation. It solved the problem of score sparsity to some extent. At the same time, this algorithm integrated closely distributed computing platform, which could make full use of the advantages of distributed platform to design and implement an increment iterative model of recommendation algorithm by using the Spark streaming and distributed storage structure. It solved the problem that collaborative filtering algorithm was hard to expand and made poor real-time performance. The experimental data used UCI public data set named MovieLens and NetFlix films’ score. The experimental results show that the improved algorithm has a good performance and makes great progress in personalized recommendation, accuracy and scalability compared with the previous algorithms. It provides a feasible algorithm integration scheme for the application of the recommended system. [ABSTRACT FROM AUTHOR]

Copyright of Computer Measurement & Control is the property of Magazine Agency of Computer Measurement & Control and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Due to the high computational complexity of massive MIMO system,a low-complexity sparse channel estimation algorithm was proposed utilizing the inherent sparsity of the wireless communication channel to improve the estimation performance.The proposed algorithm separated channel tap from noise space based on the traditional discrete Fourier transform by adopting integral separation algorithm.This channel estimation algorithm need only calculate the channel tap,thus markedly reducing complexity of the algorithm.Numerical simulations show that proposed algorithm can approach to the performance of the minimum mean-square error estimator while maintaining lower complexity.

To realize multi-dimensional dynamic spectrum access, an approximate model was proposed for the global power spectral density (PSD)of primary users (PU). Based on the proposed model, a novel cooperative spectrum sensing algorithm was proposed, and its overall flow was also built to obtain global information in the network of PU. The global information included locations, occupied frequency bands and transmitting powers of the PU. Then, an estimator of mod-el coefficient vector was designed by utilizing the th of variational Bayesian inference (VBI). Simulation results show that the proposed approximate model has good accuracy, and the corresponding estimation algorithm of model coefficient vector has good convergence and stability. Meanwhile, the relationship between SNR and the leakage of ag-gregate spurious power (LASP)was pointed out, and the influence of SNR and LASP on MSE performance was also discussed. Furthermore, it is proved that the proposed algorithm has better MSE performance than another algorithm since the sparsity of model coefficient vector is util zed.

While the use of SAR Tomography (TomoSAR) based on Compressive Sensing (CS) makes it possible to reconstruct the height profile of an observed scene, the performance of the reconstruction decreases for a structural observed scene. To deal with this issue, we propose using TomoSAR based on Block Compressive Sensing (BCS), which changes the reconstruction of the structural observed scene into a BCS problem under the principles of CS. Further, the block size is established by utilizing the relationship between the characteristics of the structural observed scene and the SAR parameters, such that the BCS problem is efficiently solved with a block sparse l1/l2 norm optimization signal model. Compared with existing CSTomoSAR methods, the proposed BCS-TomoSAR method makes better use of the sparsity and structure information of a structural observed scene, and has higher precision and better reconstruction performance. We used simulations and Radarsat-2 data to verify the effectiveness of this proposed method.

As the borrowing rate at libraries declines significantly year by year, the book recommendation system becomes increasingly important in that it can help libraries to promote the borrowing and utilization of books. Collaborative filtering is one of the most successful and widely used technologies for recommendation systems. It leverages the similarity in reading taste between readers as the basis of recommendation. However, this method can’t directly apply to book recommendation in academic libraries because their collection policy, reader’s behavior, and business model are different from business bookstores. In particular, the distributions of borrowed books at academic libraries are sparse and imbalanced, which seriously affect the accuracy and quality of book recommendation. This paper proposes a new method for book recommendation at academic libraries. The proposed method incorporates multi-level association rule exploration algorithm and taxonomy tree to address the issues of sparsity and imbalance. In addition, we present a best compromise solution for selection of recommended books. The feedback from implicit indicator information is further used to improve the accuracy and effectiveness of recommendation. Experimental results show that the proposed method can handle the sparsity and imbalance in transactions and discover the books of the reader’s potential interest, which validates the utility and effectiveness of recommendation. [ABSTRACT FROM AUTHOR]

While collaborative filtering(CF) algorithm shave been widely applied in recommender systems, the sparsity of the target rating data issue was still a crucial bottleneck for most existing CF methods. To this end, this paper proposed a novel CF algorithm-----LMTCF, to address the sparse collaborative filtering problem. In certain optimal latent subspace, LMTCF aimed to reduce the sparsity in target data by transferring knowledge (also called as the latent factors of users and items) from multiple dense auxiliary data sources as well as preserving the local geometrical structure of the target data. Besides, LMTCF could additionally address two key issues effectively, i. e. , negative transfer and inadequate transfer learning, thus allowing more positive knowledge transferred across domains to reduce the sparsity of target data. Experiments on two benchmark datasets demonstrate that this method significantly outperforms state-of-the-art CF methods. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Signal Processing is the property of Journal of Signal Processing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

For new users with a small number of rating.it is difficult to find the approximate neighbor set of the target user by the traditional method. This paper proposed a collaborative filtering algorithm based on conditional walk bipartite graph. Firstly, it transformed the user-project scoring matrix into user-project bipartite graphs according to the bipartite graph network of complex network theory. It used the conditional walk to compute the similarity between the target user and other users.and then scored unrated items according to the collaborative filtering algorithm and generated recommendations. 'Hie results show that under the same data sparsity condition. the co-filtering algorithm based on conditional bipartite map is superior to the other twro traditional algorithms in MAE and accuracy, and the algorithm's recommendation accuracy is improved. The higher the degree of data sparsity, the higher the recommendation quality of the recommendation algorithm based on conditional biped graph. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Image & Graphics is the property of Editorial Office of Journal of Image & Graphics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Compressive sensing SAR imaging can significantly reduce the sampling rate and the amount of data,but it is essential only in the case where the reflection coefficients of SAR scene are sparse. This paper proposed a compressive sensing SAR imaging method based on wavelet packet sparse representation. The wavelet packet algorithm is used to choose the most sparse representation of the SAR scene by training the same type of SAR images. By solving for the minimum 1 l norm optimization, the SAR scene reflection coefficients can be reconstructed. Unambiguous SAR image can be produced with the proposed method even with fewer samples. SAR data simulation experiments demonstrate the efficiency of the proposed method.

Copyright of Transactions of Beijing Institute of Technology is the property of Beijing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Chongqing University of Posts & Telecommunications (Natural Science Edition) is the property of Chongqing University of Posts & Telecommunications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Coal Science & Technology (0253-2336) is the property of Coal Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Considering that the L1-norm multiple kernel learning （MKL） method may lead to discard useful informations and yield degenerated generalization performance when it produces sparse solution of the kernel weights, and Lp-norm （p > 1） multiple kernel learning （MKL） method may result in numerous redundant information and is sensitive to noise when the method produces the kernel weight with non-sparse solution. This paper proposed a method called generalized sparse MKL （GSMKL）method by introducing an elastic-net-type constraint on the kernel weights. More specifically, it was an MKL method with a constraint on the combination of the L1-norm and Lp-norm （p > 1） on the kernel weights, which could not only adjust the sparseness flexibly but also encourage the grouping effect on the solution. And it believed that both L1-norm MKL and Lp-norm MKL could be regarded as special cases. The mixed constraint in the method was non-linear constraint, and the method utilized second-order Taylor expansions to approximate the mixed constraint. Besides, it employed the semi-infinite program （SIP）to solve the optimization problem. Experimental results show that the improved algorithm, under the condition for the existence of dynamic adjustment sparseness, can not only achieve good classification performance, but also facilitate the grouping effect, so the improved algorithm is efficacious and feasible. [ABSTRACT FROM AUTHOR]

For the issues of tag quality and data sparsity in tag-based recommendation algorithm, this paper proposed a multi topic domain hybrid recommendation algorithm fusing similarity network. First it separated resources and tags to different topic domains to solve the ambiguity problem of tags. Then it established similarity network to find the synonymous words of target tags, and used the synonymous words to extend users' and resources' interest. At last it combined the target user's own pre-ference information and his neighbours' preference information to generate recommendation. Experimental results show that the approach has a better recommendation results. [ABSTRACT FROM AUTHOR]