Causality extraction is a kind of relation extraction task in Natural Language Processing (NLP), which mines event pairs with causality from text by constructing event graph, and play important role in applications of finance, security, biology and other fields. Firstly, the concepts such as event extraction and causality were introduced, and the evolution of mainstream methods and the common datasets of causality extraction were described. Then, the current mainstream causality extraction models were listed. Based on the detailed analysis of pipeline based models and joint extraction models, the advantages and disadvantages of various methods and models were compared. Furthermore, the experimental performance and related experimental data of the models were summarized and analyzed. Finally, the research difficulties and future key research directions of causality extraction were given.

Natural Language Generation (NLG) technologies use artificial intelligence and linguistic methods to automatically generate understandable natural language texts. The difficulty of communication between human and computer is reduced by NLG, which is widely used in machine news writing, chatbot and other fields, and has become one of the research hotspots of artificial intelligence. Firstly, the current mainstream methods and models of NLG were listed, and the advantages and disadvantages of these methods and models were compared in detail. Then, aiming at three NLG technologies:text-to-text, data-to-text and image-to-text, the application fields, existing problems and current research progresses were summarized and analyzed respectively. Furthermore, the common evaluation methods and their application scopes of the above generation technologies were described. Finally, the development trends and research difficulties of NLG technologies were given.

The existing semantic segmentation algorithms have too many parameters and huge memory usage, so that it is difficult to meet the requirements real-world applications such as automatic driving. In order to solve the problem, a novel, effective and lightweight real-time semantic segmentation algorithm based on Separable Pyramid Module (SPM) was proposed. Firstly, factorized convolution and dilated convolution were adopted in the form of a feature pyramid to construct the bottleneck structure, providing a simple but effective way to extract local and contextual information. Then, the Context Channel Attention (CCA) module based on computer vision attention was proposed to modify the channel weights of shallow feature maps by utilizing deep semantic features, thereby optimizing the segmentation results. Experimental results show that without pre-training or any additional processing, the proposed algorithm achieves mean Intersection-over-Union (mIoU) of 71.86% on Cityscapes test set at the speed of 91 Frames Per Second (FPS). Compared to Efficient Residual Factorized ConvNet (ERFNet), the proposed algorithm has the mIoU 3.86 percentage points higher, and the processing speed of 2.2 times. Compared with the latest Light-weighted Network with Efficient Reduced Non-local operation for real-time semantic segmentation (LRNNet), the proposed algorithm has the mIoU slightly lower by 0.34 percentage points, but the processing speed increased by 20 FPS. The experimental results show that the proposed algorithm has great value for completing tasks such as efficient and accurate street scene image segmentation required in automatic driving.

The classical Graph Twiddling (GT) algorithm is the MapReduce implementation of triangle parallel enumeration algorithm. However, the GT algorithm can only enumerate the triangle structure of whole graph and can not enumerate the triangle structure of candidate vertexes directly. To solve the problem, a parallel algorithm was proposed for directly enumerating the triangle structure of candidate vertexes. Firstly, the set of all the combinations of candidate vertexes for forming triangle was given by analyzing the distribution of candidate vertexes. Then, through the screening of the set, the triangle structure of candidate vertexes was directly enumerated. Finally, the proposed algorithm was implemented on Spark to achieve high efficiency and popularity. The contrast experiment was completed on artificial datasets and real datasets. The experimental results show that, compared with the GT algorithm, the running time of the proposed algorithm is only 1/3 of the running time of GT algorithm, and the running time on Spark is only 1/7 of the running time on Hadoop. The proposed algorithm can be used to generate the triangle dataset of any candidate vertex directly and efficiently.

To overcome the prematurity of Group Search Optimizer (GSO) and improve its convergence speed, a producer pre-selection mechanism based self-adaptive group search optimizer (PSAGSO) algorithm was proposed. Firstly, the reverse mutation operator and pre-selection mechanism were employed to generate a new producer by producer-scrounger model to guide the search directions of scrounger and effectively maintain the diversity of population. Secondly, a self-adaptive method based on linear decreasing weight was adopted to adjust the proportion of rangers, which is to improve individual vigor of the population and benefit to escape from local optima. Experiments were conducted on a set of 12 benchmark functions. For 30-dimensional function optimization, the test data obtained by the PSAGSO algorithm was better than that in the literature (HE S, WU Q H, SAUNDERS J R. Group search optimizer: an optimization algorithm inspired by animal searching behavior. IEEE Transactions on Evolutionary Computation, 2009, 13(5): 973-990). For 300-dimensional numerical optimization problems, the PSAGSO algorithm exhibited better performance. The experimental result demonstrates that the PSAGSO algorithm improves the group search optimizer, and to some extent it improves the algorithm convergence speed and accuracy.

The control chart is in uncertain abnormal state when the plotted-point is close to the critical value, or the number of points is close to the prescriptive target, or there is concurrence of many abnormities. The traditional methods are hard to complete the pattern recognition. Considering the concurrence of trend pattern and cycle pattern, the original control chart signal was decomposed by wavelets. The different abnormal signals were reconstructed with appropriate wavelet coefficients. By curve fitting, the goodness of fit to the reconstruction wavelets was taken as the characteristic number of abnormal pattern. Then the occurrence degrees of uncertain patterns were calculated by inputting the characteristic numbers into membership function of corresponding patterns. The simulation model of this approach was developed under Matlab/Simulik. Finally, an application example was given and the result shows the feasibility of this approach.

An echo cancellation method on digital repeater of mobile communication system was presented to solve the problem of traditional adaptive filter, which cannot eliminate the sub-path echo in complex multipath channel. Firstly, based on phase space reconstruction theory, the signal that came from the donor antenna and contained echo was reconstructed; hence, the number of sensors was not fewer than the number of sources in blind source separation. Secondly, Independent Component Analysis (ICA) algorithm was used to separate reconstructed signal. Finally, desired signal was determined by the correlation of sent signal and separated signal. In the experiment on the multi-carrier Global System of Mobile communication (GSM) source with complex multipath echo, correlation coefficient of desired signal was up to 0.9593. The echo cancellation method based on blind source separation proves to be an effective way to eliminate the echo in complex multipath channel.

In order to deal with problem in private cloud environment caused by computing tasks with large amount of data, intensive computing and complex processing, an implementation of private cloud system based on dynamic service was proposed on the basis of public cloud computing and the characteristics of private cloud environment, which was able to adapt large-scale data processing. In this implementation, computing tasks were described by job files, processing workflows were constructed dynamically by job logic, service requests were driven by data streams and the large-scale data processing could be reflected more efficiently in MapReduce parallel framework. The experimental results show that this implementation offers a high practical value, can deal with computing tasks with large amount of data, intensive computing and complex processing correctly and efficiently.

An PCA-based location algorithm of human face features was presented.Eigenmouth was generated by prinicipal component analysis(PCA) with labelled mouth region image set.For an input image,the mouth region was recognized according to least residual error energy(LREE) criteria after the face region was first located.Then a restricted window scan stategy and comparability mearsurement P were introduced into the LREE algorithm.Experiment results show that the proposed method can locate mouth region easily,accurately and robustly.