Search

A semantic-extension-based algorithm for short texts is proposed, by involving the Word2vec and the LDA model, to improve the performance of classification, which is frequently deteriorated by semantic dependencies and scarcity of features. For every keyword within a short text, weighted synonyms and related words can be generated by the Word2Vec and LDA model, respectively, and subsequently be inserted to extend the short text to a reasonable length. We not only have established a criterion by means of similarity estimation to determine whether a sentence should be extended, we designed a scheme to choose the number of extended words. The extended text will be classified. Experimental results show that, the classification performance of the proposed algorithm, in terms of the precision rate, is approximately 5% higher than that of the TF-IDF model and approximately 10% higher than that of the VSM method.

To protect the copyright of digital image, this paper proposed a combined Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT) based watermarking scheme. To embed the watermark, the cover image was decomposed by a 2-level DWT, and the HL2 sub-band coefficient was divided into 4x4 blocks, then the DCT was performed on each of these blocks. The watermark bit was embedded by predefined pattern_0 or pattern_1 on the middle band coefficients of DCT. After watermark insertion, inverse DCT was applied to each of the 4x4 blocks of HL2 sub-band coefficient, and inverse DWT was applied to obtain the watermarked image. For watermark extraction, the watermarked image, which may be attacked by various image attacks, was decomposed with 2-level DWT and DCT similarly as watermark embedding process, then correlation between middle band coefficients of block DCT and the predefined pattern (pattern_0 and pattern_1) was calculated to decide whether a bit 0 or a bit 1 was embedded. Genetic algorithm was used for embedding and extraction parameters optimization. Optimization is to maximize PSNR of the watermarked image and NCC of the extracted watermark. Experiment results show that the proposed scheme in this paper is robust against many image attacks, and improvement can be observed when compared to other existing schemes.

According to the features of DCT-based and DFT-based video watermark algorithms against attacks and considering 3D video structural characteristic, a new robust blind video watermark algorithm in transform domain combining with 3D video correlation is proposed. DCT-based watermark and DFT-based watermark are embedded in left and right images of a video frame using opposite embedding rules, and the watermark can be extracted by integrating 3D video correlation, and more robust watermark is extracted from the DCT-based watermark and the DFT-based watermark by comparing their performance. The algorithm can significantly improve the robustness of 3D video watermark against resolution attack and coding attack. The experimental results show that the proposed algorithm has excellent performance against resolution attack and coding attack, and is obviously better than DCT-based and DFT-based algorithms.

Peer-to-Peer (P2P) applications have overtaken web stream as the most significant portions on the high-speed network, so P2P applications identification is important to a broad range of network operations. By deriving the transport/network layer headers of the packets of TCP flow, we obtained some attributes of all kinds of P2P traffic without relying on packet payload and port number, which leaded to a novel method for P2P traffic identification based on support vector machine (SVM). The method only needed to deal with the TCP packet of SYN and SYN+ACK flags. Experiment results show this method classifies the traffic achieved the high accuracy.

Integrity measurement is just one of the most important features in a trusted computing platform to provide evidence of platform configuration and behavior. Each set of software instructions (binary code) is measured and recorded before it is executed. Especially, all of boot components which be recorded in pre-boot stage can’t be re-measured until the platform is rebooted. However, the reality is that many components belong to pre-boot stage, such as BIOS, Boot Loader and MBR, are easily modified or updated in runtime environment. If these updated results of components are treated as configuration requirements for a remote server-side to verify the access request, the remote network access connection will fail. It must restart the host to re-measure there components, and the result of the new measure is not guaranteed right absolutely. We solve this problem by embedding Integrity Measurement Agency Component (IMAC) which has the capability of monitoring transformation, verifying credibility, updating Platform Configuration Registers and recording integrity measurement logs in Linux kernel. The result of performance analysis demonstrates that the method is feasible and credible.

To resolve the problem that the peak-to-average power ratio (PAPR) in orthogonal frequency-division multiplexing (OFDM) systems is too high, an improved selected mapping (SLM) is proposed in this paper. By scaling the selected signal amplitude and dividing the OFDM vector in the transmitter, it saves the system bandwidth and reduce the PAPR efficiently, and it doesn’t need side information to be sent. At the same time, the random phase sequence information is rapidly recovered by using related detective algorithm, which largely raises the detective efficiency in the receiver. The simulation results show that the improved SLM could reduce the PAPR more efficiently while the bit error rate (BER) is similar to the one of the typical SLM.

Expression of blind message in the optimal-type blind signature scheme was modified so that it can combine with threshold signature, which leads to an effective threshold blind signature scheme based on elliptic curve cryptography, it has the following securities: strong verifiability, nonrepudiation, unforgeability, blindness and robustness, and also takes advantage of characteristics of elliptic curve such as higher security and shorter key, as well as lower computational complexity compared with the previous schemes.