International Journal of Computational Intelligence and Applications

Tumor classification based on microarray gene expression data is easy to fall into overfitting because such data are composed of many irrelevant, redundant, and noisy genes. Traditional gene selection methods cannot achieve satisfactory classification results. In this study, we propose a novel multi-target hybrid gene selection method named RMOGA (ReliefF Multi-Objective Genetic Algorithm), which aims to select a few genes and obtain good tumor recognition accuracy. RMOGA consists of two phases. Firstly, ReliefF is used to select the top 5% subset of genes from the original datasets. Secondly, a multi-objective genetic algorithm searches for the optimal gene subset from the gene subset obtained by the ReliefF method. To verify the validity of RMOGA, we conducted extensive experiments on 11 available microarray datasets and compared the proposed method with other previous methods. Two classical classifiers including Naive Bayes and Support Vector Machine were used to measure the classification performance of all comparison methods. Experimental results show that the RMOGA algorithm can yield significantly better results than previous state-of-the-art methods in terms of classification accuracy and the number of selected genes.

In this paper, we proposed a deep fusion model for telephone comments recognition, named CNN-BiGRU. Traditionally, the most used algorithms in text classification are Convolutional Neural Network (CNN), Long and Short Term Memory (LSTM) and Bi-Gated Recurrent Neural Network (BiGRU). For CNN, it can extract the feature form the neighbors, and a softmax layer is followed for classification. The global feature is not included in the CNN model. LSTM introduces the gate, which can capture the information before the node. BiGRU is developed from LSTM, and it can find the features in the context. So compared to LSTM, BiGRU not only includes the information before, but also can capture the following features. Thus, LSTM and BiGRU can extract the global features, but cannot capture the local features. In order to deal with this weakness, we proposed a fusion model for comments classification, which combines the CNN and BiGRU in our model. Different from other methods, CNN and BiGRU are parallelly connected. CNN model can extract the local feature, and BiGRU can find the global feature. Then we concatenate the two kinds of features and feed to recognition layer for classification. Then we use our model to classify the telephone comments; compared with the traditional machine SVM and tow deep neural models — CNN and BiGRU — our model performed better.

With the appearance of complete mammalian genomes, comparative approaches have experienced a recent upsurge. Searching maximal exact match is among the most used tasks in sequence searching within a larger DNA sequence or database. Many exact algorithms have been designed to deal with this problem. The best improvements made by these algorithms have led to a time and space complexity of O(n) and they remain practically less effective for large sequences. Heuristic methods will therefore be a good alternative to implement. In this work, we present an efficient heuristic algorithm based on PSO metaheuristic to deal with the problem of searching the maximal exact match of small searched sequences in large ones. The time and space complexity of the designed algorithm is of O(1). The experimental results showed the efficiency of the proposed search algorithm for finding maximal exact match in large sequences when compared with two other sequence searching algorithms.

As the popularity of wireless networks deepens, the diversity of device types and hardware environments makes network data take on heterogeneous forms while the threat of malicious attacks from outside can prevent ordinary methods from mining information from abnormal data. In view of this, the research will be devoted to the feature processing of the anomalous data itself, and the convolutional operation of the anomalous information by the convolutional neural network (CNN). This is to extract the internal information. In the first step of the cascaded CNNs, the dimensions of the anomaly data will be processed, the anomaly data will be sorted under the concept of relevance grouping, and then the sorted results will be added to the convolution and pooling. The performance test uses three datasets with different feature capacities as the attack sources, and the results show a 13.22% improvement in information mining performance compared to the standard CNN. The extended CNN step will perform feature identification for homologous or similar network threats, with feature expansion within the convolutional layer first, and then pooling to reduce the computational cost. The test results show that when the maximum value domain of linear expansion is 2, the model has the best feature recognition performance, fluctuating around 85%; The model comparison test results show that the accuracy of the extended CNN is higher than that of the standard CNN, and the model stability is better than that of the back propagation (BP) neural network. This indicates that the cascaded CNN dual module can mine for the data itself, thus ignoring the risk unknowns, and this connected CNN has some practical significance. The proposed cascaded CNN module applies advanced neural network technology to identify internal and external risk data. The research content has important reference value for the security management of IoT systems.

Food price forecasts in the agricultural sector have always been a vital matter to a wide variety of market participants. In this work, we approach this forecast problem for the weekly wholesale food price index in the Chinese market during a 10-year period of January 1, 2010–January 3, 2020. To facilitate the analysis, we propose the use of the nonlinear auto-regressive neural network. Technically, we investigate forecast performance, based upon the relative root mean square error (RRMSE) as the evaluation metrics, corresponding to one hundred and twenty settings that cover different algorithms for model estimations, numbers of hidden neurons and delays, and ratios for splitting the data. Our experimental result suggests the construction of the neural network with three delays and 10 hidden neurons, which is trained through the Levenberg–Marquardt algorithm, as the forecast model. It leads to high accuracy and stabilities with the RRMSEs of 1.93% for the training phase, 2.16% for the validation phase, and 1.95% for the testing phase. Comparisons of forecast accuracy between the proposed model and some other machine learning models, as well as traditional time-series econometric models, suggest that our proposed model leads to statistically significant better performance. Our results could benefit different forecast users, such as policymakers and various market participants, in policy analysis and market assessments.

In recent years, the rapid development of mobile devices and embedded system raises a demand for intelligent models to address increasingly complicated problems. However, the complexity of the structure and extensive parameters press significantly on efficiency, storage space, and energy consumption. Additionally, the explosive growth of tasks with enormous model structures and parameters makes it impossible to compress models manually. Thus, a standardized and effective model compression solution achieving lightweight neural networks is established as an urgent demand by the industry. Accordingly, Dynamic Channel Ranking Strategy (DCRS) method is proposed to compress deep convolutional neural networks. DCRS selects channels with high contribution of each prunable layer according to compression ratio searched by reinforcement learning agent. Compared with current model compression methods, DCRS efficaciously applies various channel ranking strategies on prunable layers. Experiments indicate with a 50% compression ratio, compressed MobileNet achieved 70.62% top1 and 88.2% top5 accuracy on ImageNet, and compressed ResNet achieved 92.03% accuracy on CIFAR-10. DCRS reduces more FLOPS in these neural networks. The compressed model achieves the best Top-1 and Top-5 accuracy on ResNet50, the best Top-1 accuracy on MobilNetV1.

In recent years, rainfall prediction has received major attention in research areas because of its demanding applications in pollution control management and flood control management. Despite having numerous learning-based approaches to calculate future rainfall trends, it remains inefficient to predict rainfall occurrences by learning linear and nonlinear data patterns of historical weather information (i.e., exact prediction value is complicated to be predicted). These complications are addressed with the evolution of stochastic models which have a greater ability to minimize prediction bias and represent long-term weather variability. Therefore, this paper proposes a novel modified stochastic fuzzy Aquila (MSFA) algorithm to make precise predictions regarding future trends by evaluating rainfall time series data. The proposed MSFA algorithm is applied in rainfall prediction applications in evaluating the effectiveness of the proposed stochastic model. Here, 10 features of the open weather dataset collected from Tamil Nadu are provided as input for the proposed rainfall prediction design. The data inconsistencies such as undesirable format and missing values are structured using preprocessing procedures, namely data arrangement, null value removal, and data partitioning. The preprocessed data are fed into the proposed MSFA algorithm which learns the data features more precisely and predicts the probable occurrence of rainfall. To evaluate the performances of the proposed MSFA algorithm, the metrics such as mean absolute error (MAE), coefficient of determination, root mean squared logarithmic error (RMSLE), and root mean square error (RMSE) are analyzed. The experimental results illustrate that the proposed MSFA algorithm achieves superior performance in terms of all metrics.

The effectiveness of machine learning approaches depends on the quality of the data representation. In fact, some representations may mislead such learning approaches upon concealing relevant explanatory variables. Although feature engineering, that utilizes domain knowledge and/or expert supervision, yields typical data representation techniques, generic unsupervised feature learning represents an even more objective alternative to determine relevant attributes and generate optimal feature spaces. In this paper, we propose a new fuzzy unsupervised feature learning approach (FUL) that automatically derives new features by revealing the intrinsic structure of the data. In fact, FUL exploits the clusters and the associated fuzzy memberships generated by a fuzzy C-means algorithm, and devises new basis functions and their corresponding representation. The experiments results showed that FUL overtakes relevant state of the art approaches. It yielded the highest F1-measure with an improvement of 8%, 11%, 3%, and 4% on Parkinson, Epilepsy, Gait, and breast cancer datasets, respectively.