This paper proposes a C-RNN forecasting method for Forex time series data based on deep-Recurrent Neural Network (RNN) and deep Convolutional Neural Network (CNN), which can further improve the prediction accuracy of ...
详细信息
This paper proposes a C-RNN forecasting method for Forex time series data based on deep-Recurrent Neural Network (RNN) and deep Convolutional Neural Network (CNN), which can further improve the prediction accuracy of deep learning algorithm for the time series data of exchange rate. We fully exploit the spatio-temporal characteristics of forex time series data based on the data-driven method. On the exchange rate data of nine major foreign exchange currencies, the experimental comparison of the forecasting method shows that the C-RNN foreign exchange time series data prediction method constructed in this paper has better applicability and higher accuracy.
Clustering analysis has been widely used in pattern recognition and image processing in recent years, which is an important research field of data mining. Data publishing in social networks is threatened by the leakag...
详细信息
Clustering analysis has been widely used in pattern recognition and image processing in recent years, which is an important research field of data mining. Data publishing in social networks is threatened by the leakage of private information nowadays. This paper proposes a privacy preservation scheme of sensitive data publishing in social networks based on Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) algorithm to tackle this issue. The scheme is divided into an online process and an offline process. Specifically, we present the Maximum Delay Anonymous Clustering Feature (MDACF) tree data publishing algorithm.
Multi-task learning (MTL) is an important subject in machine learning and artificial intelligence. Its applications to computer vision, signal processing, and speech recognition are ubiquitous. Although this subject h...
详细信息
Nowadays, with the increasing number of protein sequences all over the world, more and more people are paying their attention to predicting protein subcellular location. Since wet experiment is costly and time-consumi...
详细信息
Nowadays, with the increasing number of protein sequences all over the world, more and more people are paying their attention to predicting protein subcellular location. Since wet experiment is costly and time-consuming, the automatic computational methods are urgent. In this paper, we propose a variant model based on Long Short-Term Memory(LSTM) to predict protein subcellular location. In this model, we employ LSTM to capture long distance dependency features of the sequence data. Moreover, we adjust the loss function of the loss layer to solve multi-label classification problem. Experimental results demonstrate that, compared with the traditional machine learning methods, our method achieves the best performance in various evaluation metrics.
By constructing a list of IF-THEN rules, the traditional ant colony optimization(ACO) has been successfully applied on data classification with not only a promising accuracy but also a user comprehensibility. However,...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
By constructing a list of IF-THEN rules, the traditional ant colony optimization(ACO) has been successfully applied on data classification with not only a promising accuracy but also a user comprehensibility. However, as the collected data to be classified usually contain large volumes and redundant features, it is challenging to further improve the classification accuracy and meanwhile reduce the computational time for *** paper proposes a novel hybrid mutual information based ant colony algorithm(mrAM+) for classification. First, a maximum relevance minimum redundancy feature selection method is used to select the most informative and discriminative attributes in a dataset. Then, we use the enhanced ACO classifier(i.e., AM+)to perform the classification. Experimental results show that the proposed mrAM+ outperforms other seven state-of-art related classification algorithms in terms of accuracy and the size of model.
This paper aims at developing a clustering approach with spectral images directly from CASSI compressive measurements. The proposed clustering method first assumes that compressed measurements lie in the union of mult...
详细信息
Recently the research of location privacy preserving has become a hot spot. Location privacy preserving involves not only a single location privacy but also the trajectory privacy where the mobile users in different l...
详细信息
Recently the research of location privacy preserving has become a hot spot. Location privacy preserving involves not only a single location privacy but also the trajectory privacy where the mobile users in different locations publish the consecutive query requests. In this paper, we consider the problem of trajectory privacy preserving in MSN. Particularly, we propose privacy preserving algorithms based on R-constraint dummy trajectory (RcDT). By constraining the generating range R of the dummy positions, the generated dummy positions are within a certain range near the real locations. Furthermore, the dummy trajectories which have higher similarity to the real trajectories are generated via constraining both the single location exposure risk and the trajectory exposure risk.
Most existing approaches of learning to rank treat the effectiveness of each query equally which results in a relatively lower ratio of queries with high effectiveness (i.e. rich queries) in the produced ranking model...
详细信息
With the development of ordering platform, an increasing number of people are paying their attention to design a suitable recommender system. Most of the traditional recommender systems are based on the abundant ratin...
详细信息
With the development of ordering platform, an increasing number of people are paying their attention to design a suitable recommender system. Most of the traditional recommender systems are based on the abundant rating information of users. However, Only historical order data can be provided to the recommender system in ordering platform as training data. This paper proposes an improved Collaborative Filtering algorithm based on historical order data of restaurants. The recommender system includes two parts: 1) rule generation module, we define a new method for measuring the similarity between dishes. Furthermore, we incorporate an incremental learning method in this module. 2) recommendation module, we design user interest vector and propose a noise filtering method. Experimental results demonstrate that the proposed algorithm can effectively improve the performance of recommendation in terms of the accuracy and coverage ratio. Moreover, our recommender system has been successfully put into service.
Effectiveness is the most important factor considered in the ranking models yielded by algorithms of learning to rank (LTR). Most of the related ranking models only focus on improving the average effectiveness but ign...
详细信息
暂无评论