Heterogeneous Feature Analysis on Twitter Data Set for Identification of Spam Messages

Spam is an undesirable content that present on online social networking sites, while spammers are the users who post this content on social networking sites. Unwanted messages posted on Twitter may have several goals and the spam tweets can interfere with statistics presented by Twitter mining tools and squander users’ attention.. Since Twitter has achieved a lot of attractiveness through-out the world, the interest towards it by the spammers and malevolent users is also increases. To overcome the spam problems many researchers proposed ideas using machine learning algorithms for the identification of spam messages. Not only the selection of classifiers but also the variegated feature analysis is essential for the identification of irrelevant messages in social networks. The proposed model performs a heterogeneous feature analysis on the twitter data streams for classifying the unsolicited messages using binary and continuous feature extraction with sentiment analysis on social network datasets. The features created are assessed using significant stratagems and the finest features are selected. A classifier model is built using these feature vectors to predict and identify the spam messages in Twitter. The experimental results clearly show that the proposed Sentiment Analysis based Binary and Continuous Feature Extraction model with Random Forest (SA-BC-RF) approach classifies the spam messages from the social networks with an accuracy of 90.72% when compared with the other state-of-the-art methods.

Joint Feature-Space and Sample-Space Based Heterogeneous Feature Transfer Method for Object Recognition Using Remote Sensing Images with Different Spatial Resolutions

To improve the classification results of high-resolution remote sensing images (RSIs), it is necessary to use feature transfer methods to mine the relevant information between high-resolution RSIs and low-resolution RSIs to train the classifiers together. Most of the existing feature transfer methods can only handle homogeneous data (i.e., data with the same dimension) and are susceptible to the quality of the RSIs, while RSIs with different resolutions present different feature dimensions and samples obtained from illumination conditions. To obtain effective classification results, unlike existing methods that focus only on the projection transformation in feature space, a joint feature-space and sample-space heterogeneous feature transfer (JFSSS-HFT) method is proposed to simultaneously process heterogeneous multi-resolution images in feature space using projection matrices of different dimensions and reduce the impact of outliers by adaptive weight factors in the sample space simultaneously to reduce the occurrence of negative transfer. Moreover, the maximum interclass variance term is embedded to improve the discriminant ability of the transferred features. To solve the optimization problem of JFSSS-HFT, the alternating-direction method of multipliers (ADMM) is introduced to alternatively optimize the parameters of JFSSS-HFT. Using different types of ship patches and airplane patches with different resolutions, the experimental results show that the proposed JFSSS-HFT obtains better classification results than the typical feature transferred methods.

MDNN: A Multimodal Deep Neural Network for Predicting Drug-Drug Interaction Events

The interaction of multiple drugs could lead to serious events, which causes injuries and huge medical costs. Accurate prediction of drug-drug interaction (DDI) events can help clinicians make effective decisions and establish appropriate therapy programs. Recently, many AI-based techniques have been proposed for predicting DDI associated events. However, most existing methods pay less attention to the potential correlations between DDI events and other multimodal data such as targets and enzymes. To address this problem, we propose a Multimodal Deep Neural Network (MDNN) for DDI events prediction. In MDNN, we design a two-pathway framework including drug knowledge graph (DKG) based pathway and heterogeneous feature (HF) based pathway to obtain drug multimodal representations. Finally, a multimodal fusion neural layer is designed to explore the complementary among the drug multimodal representations. We conduct extensive experiments on real-world dataset. The results show that MDNN can accurately predict DDI events and outperform the state-of-the-art models.

Bi-directional Coupling of an Unstructured Triangular Meshes-based Integrated Hydrodynamic Model for Heterogeneous Feature-based Urban Flood Simulation

Urban pluvial flash floods have become a matter of widespread concern, as they severely impact people’s lives in urban areas. Hydrological and hydraulic models have been widely used for urban flood management and urban planning. Traditionally, to reduce the complexity of urban flood modelling and simulations, simplification or generalization methods have been used; for example, some models focus on the simulation of overland water flow, and some models focus on the simulation of the water flow in sewer systems. However, the water flow of urban floods includes both overland flow and sewer system flow. The overland flow processes are impacted by many different geographical features in what is an extremely spatially heterogeneous environment. Therefore, this article is based on two widely used models (SWMM and ANUGA) that are coupled to develop a bi-directional method of simulating water flow processes in urban areas. The open source overland flow model uses the unstructured triangular as the spatial discretization scheme. The unstructured triangular-based hydraulic model can be better used to capture the spatial heterogeneity of the urban surfaces. So, the unstructured triangular-based model is an essential condition for heterogeneous feature-based urban flood simulation. The experiments indicate that the proposed coupled model in this article can accurately depict surface waterlogged areas and that the heterogeneous feature-based urban flood model can be used to determine different types of urban flow processes.