Search

Recognition of emotion from a person's voice conveys the state of a person. There is different state of emotions. Happy, sad, angry and neutral are the primary emotions of a human being are analyzed by speech patterns. This pattern also used in the medical field for the research to know the mental state, BP and other research purpose. Nowadays robot knows the feelings of a person. In this paper the classification of human emotion based on machine learning is proposed. The proposed algorithm uses audio files which includes spoken words of different emotions. Those emotions are processed with the algorithm designed and their features are extracted. Those features are trained and recognized using SVM. [ABSTRACT FROM AUTHOR]

The process of gleaning important facts and insights from vast amounts of unstructured textual data is called text mining, sometimes referred to as text analytics. In today's data-driven business climate, text analytics is becoming increasingly significant. A key component of text mining is term-weighting, which aids in locating significant keywords or phrases in a text document. Text mining algorithms can more precisely identify and analyze documents by giving these phrases weights, which can result in important insights and knowledge discovery. The study article thoroughly analyses several methods put out in the literature to help computers detect language and text. It covers the mathematical underpinnings of term weighting schemes, categorizes them broadly into supervised and statistical methods, and provides examples from both groups. The Vector-Space Model and its variations, which are at the core of many other techniques covered in the paper, are given special emphasis in this article. The significant research highlights the necessity for machines to effectively understand language and material without engaging in plagiarism or other immoral activities. Overall, the study lays the framework for further research in this field by offering a thorough overview of contemporary language and text recognition methods. [ABSTRACT FROM AUTHOR]

Despite the advantages of cloud computing, such as its scalability and accessibility, the security of the data it holds remains a major concern. This paper proposes a hybrid structure that combines the features of two well-known encryption algorithms. The hybrid structure adopts the DES' flexibility and efficiency while also taking advantage of AES' robustness. Our goal is to offer a higher level of protection for sensitive data that is stored in cloud computing. The paper presents an overview of the various aspects of the proposed architecture, including key management, decryption, and encryption. The first step in the encryption process is to implement the AES before adding the DES for security. The paper provides an overview of the various distribution and key generation techniques. They are designed to ensure that the data is protected from unauthorized access. In addition, we discuss the decryption process, which enables users to retrieve the original data. We performed extensive tests to analyze the security and performance of the proposed architecture. These tests included comparisons with the standalone DES and AES algorithms. The results of the tests revealed that the proposed architecture performed better than the standalone DES and AES algorithms in several aspects. In addition, it exhibited better decryption and encryption speeds. The hybrid structure's improved resource utilization also leads to lower system memory and CPU utilization. Its high throughput helps improve system responsiveness and data processing. The proposed architecture ensures the security of sensitive data by minimizing the time it takes to generate and manage key generation. It also helps improve the efficiency of the process by reducing the file size overhead. The findings of the study highlight the advantages of implementing the hybrid structure in protecting sensitive data stored in cloud computing, as well as its viability. The combination of the DES and the AES offers a flexible and robust approach to safeguarding information. This report serves as a valuable contribution to the field of data security by providing a practical solution to the cloud computing security issues. [ABSTRACT FROM AUTHOR]

Object detection is an essential component of computer vision that is used in a wide variety of real-world applications, including robots, autonomous vehicles, and video surveillance. Deep learning models like Fast R-CNN, SSD, and YOLOv2 have demonstrated exceptional performance in object identification as deep learning has advanced. These models are part of the deep learning category. The detection of objects in an image is accomplished by these models by the application of a variety of methodologies and architectural frameworks. Some of them are region proposal, anchor boxes, and multi-scale feature maps. In this study, we assess the performance of various models based on the PASCAL VOC dataset and explore the layer design, setup, and evaluation metrics of each model. Using the PASCAL VOC 2007 and PASCAL VOC 2012 datasets, we discover that SSD obtains the maximum mAP, although YOLOv2 has the quickest inference performance. It's possible that Fast R-CNN is the best option for striking a balance between speed and accuracy. While selecting an object identification model, it is essential, however, to take into account the particular needs of the job as well as the features of the dataset. In general, object detection has a lot of promise to help enhance a lot of different applications that are used in the real world, and it will continue to be a hot study topic in the field of computer vision. [ABSTRACT FROM AUTHOR]