<span>The imbalanced data problems in data mining are common nowadays, which occur due to skewed nature of data. These problems impact the classification process negatively in machine learning process. In such problems, classes have different ratios of specimens in which a large number of specimens belong to one class and the other class has fewer specimens that is usually an essential class, but unfortunately misclassified by many classifiers. So far, significant research is performed to address the imbalanced data problems by implementing different techniques and approaches. In this research, a comprehensive survey is performed to identify the challenges of handling imbalanced class problems during classification process using machine learning algorithms. We discuss the issues of classifiers which endorse bias for majority class and ignore the minority class. Furthermore, the viable solutions and potential future directions are provided to handle the problems<em>.</em></span>

Text classification has become very serious problem for big organization to manage the large amount of online data and has been extensively applied in the tasks of Natural Language Processing (NLP). Text classification can support users to excellently manage and exploit meaningful information require to be classified into various categories for further use. In order to best classify texts, our research efforts to develop a deep learning approach which obtains superior performance in text classification than other RNNs approaches. However, the main problem in text classification is how to enhance the classification accuracy and the sparsity of the data semantics sensitivity to context often hinders the classification performance of texts. In order to overcome the weakness, in this paper we proposed unified structure to investigate the effects of word embedding and Gated Recurrent Unit (GRU) for text classification on two benchmark datasets included (Google snippets and TREC). GRU is a well-known type of recurrent neural network (RNN), which is ability of computing sequential data over its recurrent architecture. Experimentally, the semantically connected words are commonly near to each other in embedding spaces. First, words in posts are changed into vectors via word embedding technique. Then, the words sequential in sentences are fed to GRU to extract the contextual semantics between words. The experimental results showed that proposed GRU model can effectively learn the word usage in context of texts provided training data. The quantity and quality of training data significantly affected the performance. We evaluated the performance of proposed approach with traditional recurrent approaches, RNN, MV-RNN and LSTM, the proposed approach is obtained better results on two benchmark datasets in the term of accuracy and error rate.

Abstract-Security is the major concern when the sensitive information is stored and transferred across the internet where the information is no longer protected by physical boundaries. Cryptography is an essential, effective and efficient component to ensure the secure communication between the different entities by transferring unintelligible information and only the authorized recipient can be able to access the information. The right selection of cryptographic algorithm is important for secure communication that provides more security, accuracy and efficiency. In this paper, we examine the security aspects and processes involved in the design and implementation of most widely used symmetric encryption algorithms such as Data Encryption Standard (DES), Triple Data Encryption Standard (3DES), Blowfish, Advanced Encryption Standard (AES) and Hybrid Cubes Encryption Algorithm (HiSea). Furthermore, this paper evaluated and compared the performance of these encryption algorithms based on encryption and decryption time, throughput, key size, avalanche effect, memory, correlation assessment and entropy. Thus, amongst the existing cryptographic algorithm, we choose a suitable encryption algorithm based on different parameters that are best fit to the user requirements.

The modern age of technology in which most of the customer needs to wait in the supermarket for shopping because it is a highly time-consuming process. A huge crowd in the supermarket at the time of discount offers or weekends makes trouble to wait in long queues because of a barcode-based billing process. In this regard, the Internet of Things (IoT) based Smart Shopping Cart is proposed which consists of Radio Frequency Identification (RFID) sensors, Arduino microcontroller, Bluetooth module, and Mobile application. RFID sensors depend on wireless communication. One part is the RFID tag attached to each product and the other is RFID reader that reads the product information efficiently. After this, each product information shows in the Mobile application. The customer easily manages the shopping list in Mobile application according to preferences. Then shopping information sends to the server wirelessly and automatically generates billing. This experimental prototype is designed to eliminate time-consuming shopping process and quality of services issues. The proposed system can easily be implemented and tested at a commercial scale under the real scenario in the future. That is why the proposed model is more competitive as compared to others.

The idea of a smart home is getting attention for the last few years. The key challenges in a smart home are intelligent decision making, secure identification, and authentication of the IoT devices, continuous connectivity, data security, and privacy issues. The existing systems are targeting one or two of these issues whereas a smart home automation system that is not only secure but also has intelligent decision making and analytical abilities is the need of time. In this paper, we present a novel idea of a smart home that uses a machine learning algorithm (Support Vector Machine) for intelligent decision making and also uses blockchain technology to ensure identification and authentication of the IoT devices. Emerging blockchain technology plays a vital role by providing a reliable, secure, and decentralized mechanism for identification and authentication of the IoT devices used in the proposed home automation system. Moreover, the SVM classifier is applied to classify the status of devices used in the proposed smart home automation system into one of the two categories, i.e., “ON” and “OFF.” This system is based on Raspberry Pi, 5 V relay circuit, and some sensors. A mobile application is developed using the Android platform. Raspberry Pi acting as the server maintains the database of each appliance. The HTTP web interface and apache server are used for communication among the Android app and Raspberry Pi. The proposed idea is tested in the lab and real life to validate its effectiveness and usefulness. It is also ensured that the hardware and technology used in the proposed idea are cheap, easily available, and replicable. The experimental results highlight its significance and validate the proof of the concept.