Search

In the era of big data, with the continuous popularization of social media, various text data are growing in the network and in life. It is of great significance to analyze and manage text data using text classification technology. Text classification is a basic research field in the field of artificial intelligence natural language processing. Under the given criteria, it classifies text according to content. The application scenarios of text classification are very extensive, such as sentiment analysis, topic classification, relationship classification, etc. Deep learning is a method of representation learning based on data in machine learning, and it shows good classification effect in text data processing. Chinese text and English text have differences in form, sound, and image. Focusing on the uniqueness of Chinese text classification, this paper analyzes and expounds the deep learning methods used for Chinese text classification, and finally sorts out commonly used datasets for Chinese text classification. [ABSTRACT FROM AUTHOR]

The increase of IC design complexity and the continuous reduction of process feature size bring new severe challenges to static timing analysis (STA) and chip design cycle. In order to improve the efficiency of STA and shorten the chip design cycle, this paper fully considers the FinFET process characteristics and the principle of STA, and predicts the timing characteristics of another part of corners by introducing machine learning methods based on the timing characteristics of some corners. The experiment is based on an industrial design, and the results show that the proposed method uses 5 corners to predict the timing of other 31 corners, which can achieve an average absolute error of less than 2 ps, far better than the 21 process angles required by traditional methods. Thus, the proposed method significantly improves the prediction accuracy and significantly reduces the workload of static time series analysis. [ABSTRACT FROM AUTHOR]

In this paper, the effects of different task allocation strategies on the performance of sparse matrix and dense vector multiplication are discussed. It is observed that the selection of task allocation strategy can significantly affect the performance of sparse matrix, and there is no fixed task allocation strategy that can obtain the best performance for all sparse matrices. Therefore, this paper proposes an optimal task allocation strategy selection method based on machine learning. Its training process only uses sparse matrix features to characterize the input data set, and can automatically train the model for a given data set and target platform. Experiments show that, compared with the default block allocation method, the task allocation method selected by this model can achieve an average performance improvement of about 35%. [ABSTRACT FROM AUTHOR]

Aiming at improving the performance of the default loop scheduling and tile size of Pluto, a commonly used polyhedral compiler, this paper proposes a method to compute a variety of legal permutations for its default scheduling and auto-tune its performance according to the configuration space composed of permutations and tile sizes. Through the processing of scalar dimension that defines loop fusion, both intra and inter permutations for imperfect loop nest are realized. Four machine learning driven auto-tuning strategies are proposed to find the optimized combination of permutation order and tile size for a loop with a given problem size. Under the default tile size, the optimal permutation generated by the extended Pluto compiler in a parallel environment achieves a maximum speedup of 4.02 and a geometric mean of 2.12 compared with the default scheduling of Pluto. By further searching for a better combination of permutation order and tile size, the best auto-tuning strategy achieves a maximum speedup of 5.48 and a geometric mean of 2.86 compared with Pluto's default optimization in a parallel environment. In addition, the best configuration and the learned model obtained by auto-tuning for a particular problem size, when being applied to similar problem sizes, also outperform the default optimization of Pluto in various degrees. [ABSTRACT FROM AUTHOR]

Standard cell library is the foundation of chip design, analysis, and verification, and its generation requires a lot of time and server resources. Therefore, vendors often only provide standard cell libraries under a few corners. However, the design of chip performance, power consumption, and reliability requires delay information of standard cells under multiple voltages, temperatures, and parameters (such as drive strength, channel length, and threshold voltage). To quickly and accurately calculate the delay of standard cells under multiple corners, this paper proposes a machine learning-based method for multi-voltage, multi-temperature, and multi-parameter standard cell delay calculation. By studying the factors that affect the delay of standard cells in depth, data sets are extracted from the 28nm process standard cell library and timing reports. Machine learning algorithms are used to train and calibrate the standard cell delay calculation model. The establishment of the model takes only a few minutes, which is much less than the time consumed by simulation methods (usually hundreds of hours). The average calculation error of the model is 1.542 ps for unknown voltage cell delay, 1.814 ps for unknown temperature cell delay, and 2.202 ps for cell delay under different parameters. The prediction error of cell delay in the static timing analysis process is less than 3%. This method can quickly and accurately calculate the delay of standard cells in real-time and can be applied to fast timing analysis under multiple scenarios before sign-off. [ABSTRACT FROM AUTHOR]

The scale of supercomputers is expanding. Meanwhile, the complexity of scientific applications is also increasing, which leads to many job failures on supercomputers. These failed jobs causes a waste of resources and prolong the waiting time of queuing jobs, which seriously affects the reliability of the system. If these failed jobs can be predicted in advance, necessary measures can be taken to improve the system resource utilization and system execution efficiency, which is very important for the future exascale supercomputers. Therefore, this paper attempts to predict these job failures from the known traditional features and construction features, and find the features and processing methods that can reflect the user􀆳s work behavior patterns and submission behavior patterns. By combining behavior features and traditional features, a comprehensive framework based on tree structure model is proposed to predict job failure. The prediction experimental results show that the comprehensive prediction framework is better than the single model prediction, and the comparative experimental results show that the prediction effect is better than other related methods. [ABSTRACT FROM AUTHOR]

With the widespread popularity of the mobile Internet, the domestic online game market has become increasingly saturated. The cost of acquiring new users for game companies continues to increase. How to prevent the loss of existing users has become the focus of marketing. This paper predicts user churn based on a real game log data. First, user features are extracted and computed from log data. Second, a set of important features is selected by weight. Finally, a binary classification model is constructed with features as input and churn as output. 6 common algorithms such as random forest, support vector machine, multi-layer perceptron, gradient boosting decision tree, and logistic regression are comprehensively compared. The experimental results show that the random forest algorithm performs the best, and its model prediction accuracy reaches 91%. [ABSTRACT FROM AUTHOR]

Ocean mesoscale eddy is an important ocean mesoscale phenomenon that plays an important role in ocean circulation, mat erial and energy transport, and has an important impact on the safety of ship navigation and hydroacoustic communication. Efficient and accurate detection and identification of ocean mesoscale eddies are of great research value for both physical ocean cognition and ocean exploitation. Traditional eddy detection and identification methods rely on a single threshold designed by experts experiences. With the rise of deep learning, the current machine learning methods show certain advantages in the accuracy and automation of eddy detection and identification. This paper summarizes and comparatively analyzes the existing machine learning-based detection and identification methods to provide a systematic knowledge and reference basis for the development of ocean mesoscale eddy detection and identification research. [ABSTRACT FROM AUTHOR]

With the increase of transmission rate, transmission length and structure complexity of high-speed channel, channel modeling technology becomes more complex and difficult. This paper proposes a novel method by combining the popular machine learning method with high-speed channel. A large number of analog data are collected, and deep neural network (DNN) and recurrent neural network (RNN) methods are used to model the channeL Once the model is trained successfully, the eye diagram of the output signal can be predicted by the simulation model, and the signal integrity can be evaluated and analyzed quickly and accurately. In addition, in the high-speed channel, the serious interference and attenuation of the signal limits the transmission distance and transmission rate, which brings difficulties to the test and information collection. In order to recover the ideal signal, the high-speed serial link usually contains complex equalization blocks. The least mean square (LMS) algorithm is adopted to effectively eliminate the interference, reduce the bit error rate and improve the transmission rate. [ABSTRACT FROM AUTHOR]

Accurate user churn prediction ability facilitates improving user retention rate, increasing user count and increasing profitability. Most of the existing user churn prediction models are single model or simple integration of multiple models, and the advantages of multi-model integration are not fully utilized. This paper draws on the idea of Bootstrap Sampling in random forests, proposes an improved Stacking ensemble method, and applies the method to the real data set to predict the user churn. Through the experimental comparison on the validation set, the proposed method is better than the classical Stacking ensemble method with the same structure in the terms of the F1-score, recall rate and prediction accuracy of user churn. When the appropriate structure is adopted, the performance can surpass the optimal performance on the base classifier. [ABSTRACT FROM AUTHOR]

In our previous work, a big data active learning algorithm based on MapReduce was proposed. In this paper, we transplant this algorithm into the Spark environment and propose a Spark based big data active learning algorithm. Furthermore, the two algorithms are experimentally compared on four aspects: running time, number of files, number of synchronizations, and memory cost. Some valuable conclusions are obtained, which can be very helpful to researcher s in the related fields. [ABSTRACT FROM AUTHOR]

Data centers are an important auxiliary tool for business leaders to make decisions, and timely，accurate and scientific data are basic requirements and key principles. It is difficult and inefficient to find out abnormal one in huge amounts of data by human experience. In this paper, we propose an algorithm for detecting abnormal data based on machine learning. Because enterprise sales data consist of a series of relatively fixed data items, they can be recognized as a structured data sequence. Conditional Random Fields (CRFs) model is efficient for structured data sequence prediction, so it can be used as the detection model. A large number of history data are learnt and their intrinsic rules and relationship are analyzed so as to enable computers to detect abnormal data automatically. Experimental result shows the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]