Narrow Results

A new framework is introduced which allows the formulation of difficult structural classification tasks in terms of decision-theoretic-based pattern recognition. It is based on extending the classical formulation of generalized linear discriminant functions so as to permit each given object to have a different vector representation in each class. The proposed extension properly accounts for the corresponding extension of the classical learning techniques of linear discriminant functions in a way such that the convergence of the extended techniques can still be proved. The proposed framework can be considered as a hybrid methodology in which both structural and decision-theoretic pattern recognition are integrated. Furthermore, it can be considered as a means to achieve convenient tradeoffs between the inductive and deductive ways of knowledge acquisition, which can result in rendering tractable the possibly hard original inductive learning problem associated with the given task. The proposed framework and methods are illustrated through their use in two difficult structural classification tasks, showing both the appropriateness and the capability of these methods to obtain useful results.

Segmentation is one of the critical steps in historical document image analysis systems that determines the quality of the search, understanding, recognition and interpretation processes. It allows isolating the objects to be considered and separating the regions of interest (paragraphs, lines, words and characters) from other entities (figures, graphs, tables, etc.). This stage follows the thresholding, which aims to improve the quality of the document and to extract its background from its foreground, also for detecting and correcting the skew that leads to redress the document. Here, a hybrid method is proposed in order to locate words and characters in both handwritten and printed documents. Numerical results prove the robustness and the high precision of our approach applied on old degraded document images over four common datasets, in which the pair (Recall, Precision) reaches approximately 97.7% and 97.9%.

Text is a typical example of unstructured and heterogeneous data in which massive useful knowledge is embedded. Sentiment analysis is used to analyze and predict sentiment polarities of the text. This paper provides a survey and gives comparative analyses of the latest articles and techniques pertaining to lexicon-based, traditional machine learning-based, deep learning-based, and hybrid sentiment analysis approaches. These approaches have their own superiority and get the state-of-the-art results on diverse sentiment analysis tasks. Besides, a brief sentiment analysis example in the tourism domain is displayed, illustrating the entire process of sentiment analysis. Furthermore, we create a large table to compare the pros and cons of different types of approaches, and discuss some insights with respect to research trends. In addition, a lot of important sentiment analysis datasets are summarized in this survey.

Creating topic hierarchies that can be mapped to the source documents provide effective access to the required information. However, the exponential growth in the volume of information may be too overwhelming for manual approach to taxonomy building. Taxonomy software, with automated approach, but minimal human intervention, easy integration with other products, and effective taxonomy maintenance utility provide a practical and efficient solution. Many vendors serving this market segment offer a wide range of approaches and solutions, such as Semio, Autonomy, Verity, Quiver, Inxight, and SmartLogik. The aim of this study is to identify different types of taxonomy-building tools available in the market and study their features and characteristics. The selection criteria was based on the availability of an evaluation copy from the vendor along with detailed information about the products such as product manuals, white papers, and product data sheets. The findings showed that most taxonomy- building tools share common features such as hierarchical structures of the taxonomy (depth of the taxonomy), taxonomy maintenance, multiple taxonomy views (cross-referencing support), import/export utilities, taxonomy presentation and access (access points to the information), and being language independent (multilingual support). The results also revealed that visualization tools, personalization, product integration and industry-specific taxonomies were the value-added features to the taxonomy software.

Competitive markets and customers’ changing needs in the bank industry necessitate accurately predicting customers who may leave the firm in the near future. Consequently, creating an approach to predict precisely and identify churn-leading causes is a part of retention strategies in customer relationship management. The approach that has been utilized in this research to predict customer churn combines decision tree (DT) and multinomial regression (MR) to classify customers with no limitation of binary classification in the churn prediction context. A customer club dataset of a commercial bank case as a real churn problem is used in this study to benchmark the hybrid forecasting approach against its building blocks. The results showed that the hybrid forecasting approach outperformed DT and MR with an average accuracy of 87.66%, 90.74% micro-average, and 90.44% macro-average of AUC. Further analysis of the model performance per class indicated that the hybrid approach’s misclassification error for the churn class decreased significantly, which is the most costly error in churn problems. Moreover, due to the structure of hybrid forecasting approach, more interoperability is obtained by assessing the impact of features in different segments, resulting in transforming them into actionable insights. The proposed approach is applied to the banking industry to prevent financial loss by detecting leading churn causes. Accordingly, after predicting the risk of customer churn, marketers and managers can determine appropriate actions that will have the most significant retention impact on each customer by applying proactive retention marketing.

Android operating systems have grown in popularity and are currently being increasingly used on smartphone devices. Because of the quickly expanding quantity of Android malware and the potential safeguard of vast amounts of data kept on Android devices, the identification and categorization of Android malware is regarded as a big data challenge. Researchers started using deep learning (DL) techniques for malware detection, according to the most recent state-of-the-art studies. Nevertheless, researchers and practitioners face challenges such as the choice of DL architecture, extracting and processing features and evaluating results. In this paper, several traditional and hybrid machine learning (ML) models were developed and analyzed in detail to classify Android malware. First, some state-of-the-art DL model has been developed like multi-layer perceptron (MLP), deep neural network (DNN), recurrent neural network (RNN), long short-term memory (LSTM), convolutional neural network (CNN) and later several hybrid models have been developed by combining DNNs with support vector machine (SVM). Second, we have focused on maximizing achievement by fine-tuning a variety of configurations to guarantee the best possible combination of the hyper-parameters and attain the maximum statistical metric value. Finally, we have accomplished performance analysis between these hybrid models and a model based on VGG-16-CNN-1D along with SVM has been proposed to enhance the accuracy and efficacy of large-scale Android malware detection which outperforms all the other developed models using Drebin Dataset. Among the DL methods, DNN achieved the highest F1 score 99.07%. After doing a comparison between hybrid models, it was found that DNN-SVM achieved 99.12% while VGG16-CNN-1D-SVM achieved 98.56% precision, 98.78% recall and 99.20% F1 score, respectively.