Jothi K R

Bio: Jothi K R is an academic researcher from VIT University. The author has contributed to research in topics: Deep learning & SQL injection. The author has an hindex of 1, co-authored 2 publications receiving 2 citations.

An Efficient SQL Injection Detection System Using Deep Learning

Abstract: SQL Injection makes most of the applications that are based on different types of databases be it used in any devices vulnerable to cyber threat. SQL Injection is said to be one of the top most threat that database-based applications on the web. SQL Injection makes all the user‘s information present in the database vulnerable and the user‘s data may be either sold in black market or may be misused. The disadvantages of previously implemented SQLI model is that they will not know how will they be able to categorize new patterns, they will only be able to detect the patterns which they have experienced before or trained on, But our model will be able to identify whether the data entered is SQL injected or not identifying patterns in the input. The advantages to our system will be that it will be able to detect all and every type of Injection techniques. All the feature extraction and selection will be done by the model itself. Just the user should need to enter the text. It is also scalable and can extend it to a wide variety of applications. With the help of MLP model, we have achieved a cross-validated accuracy of 98% with a precision of 98% and recall of 97%.

Speech Intelligence Using Machine Learning for Aphasia Individual

Abstract: Aphasia is a neurological speech disorder which impairs the person’s ability to understand or express and also reading and writing capability of an Individual will be affected. Aphasia is caused due to stroke, brain trauma, tumor in cerebrum, brain surgery, infection in cerebrum, some neurological diseases. It also affects the language skills of an aphasia individual due to harmness in the section of cerebrum. Individuals with Aphasia face difficulties in sentence framing, analyzing meaning of words, finding correct words to express their ideas, feelings, thoughts and also problem with pronunciation and stammering, all these difficulties greatly affect their personal and professional life. This paper focus on bridging a gap between the individual with Aphasia and the receiver. This motivated us to come up with the Speech Intelligence System which act as an interface for the communication between the individual with Aphasia and the society. Based on the severity of Aphasia, our knowledge-based system will analyze the unstructured words pronounced by the speech disorder person and transformed it to the meaningful prediction Our Speech Intelligence System will interpret speech disorder person’s spoken words into text form by matching with the datasets.

Enhancing Secure Mutual Authentication in Mobile Fog Computing Environment

Abstract: Fog computing that extends the cloud nearby to the IoT devices. The advent of IoT technology generates a huge amount of data and is processed in the cloud, this increases the latency. Many time-sensitive applications are affected by delayed response. Fog computing solves the latency problem by processing near the data generating device. Because of the nature of fog computing mobility support, dynamic environment, geographic distribution, location awareness existing schemes do not provide better security. To address these issues, this paper provides mutual secure authentication between each entity in the fog like between the cloud and fog node, between fog node and gateway, between gateway and IoT device. This security mechanism improves the authentication in the fog network. Mobility-based authentication providing mechanism can be implemented with the MobFogSim mobility dataset that is collected from the Luxembourg traffic. Migration of the node is considered in mobility secure authentication.

BiDLNet: An Integrated Deep Learning Model for ECG-based Heart Disease Diagnosis

Abstract: —Every year, around 10 million people die due to heart attacks. The use of electrocardiograms (ECGs) is a vital part of diagnosing these conditions. These signals are used to collect information about the heart's rhythm. Currently, various limitations prevent the diagnosis of heart diseases. The BiDLNet model is proposed in this paper which aims to examine the capability of electrocardiogram data to diagnose heart disease. Through a combination of deep learning techniques and structural design, BiDLNet can extract two levels of features from the data. A discrete wavelet transform is a process that takes advantage of the features extracted from higher layers and then adds them to lower layers. An ensemble classification scheme is then made to combine the predictions of various deep learning models. The BiDLNet system can classify features of different types of heart disease using two classes of classification: binary and multiclass. It performed remarkably well in achieving an accuracy of 97.5% and 91.5%, respectively.

Secure Deduplication with Dynamic Updates in Multi-Tenant Cloud Environment

Abstract: Cloud storage service providers accommodate customers by offering adequate storage based on customer requirements. Multi-tenant architecture to provide low-cost resource provision. In a multitenant system single instance shares with multiple customers, similar applications data has been used multiple customers so Deduplication techniques eliminate redundant data to improve storage efficiency and Bandwidth. To improve the confidentiality of the user privacy data, convergent encryption with the deduplication techniques has been proposed to encrypt the user data before uploading it to the third-party CS Sometimes Confidentiality of the user data leads to issues with this deduplication method. To overcome these issues this paper proposed dynamic ownership management of server-side deduplication with the blowfish encryption algorithm. In this model to support tenant block-level deduplication. The proposed method computational storage efficiency and security level.

A Systematic Review of Machine Learning based Automatic Speech Assessment System to Evaluate Speech Impairment

Abstract: Aphasia is a communication disability that falls under the category of neurological speech disorder The main cause of the aphasia individual speech impairment damages certain portion of the cerebrum The aphasic speech impairment ranges from mild level to the extreme level of severity in which the individual will not be able to communicate The aphasia individuals face difficulties on the different communication categories like some individual have difficulty in formation of the clear and meaningful sentence, some have problem with understanding and some have trouble in the reading process The aphasic speech impairment experience will be unique for each individual It purely depends on the portion of the impairment in the cerebrum, position of the impairment in brain, degree of the severity and also individual age factor This research work focuses on the assessment of speech impairment in the aphasia patients, basically in order to evaluate the communication aptitude of the aphasia individual The assessment approach is based on the analysis of factors related to speech such as articulation, phonation, prosody and intelligibility Through the assessment approach, degree of the severity level of patient will be identified through the automatic speech recognition (ASR) methodologies This work helps the medical examiners namely neurologists and speech therapists to perform the effective speech analysis for the individuals with aphasia speech disorder

Input Validation Vulnerabilities in Web Applications: Systematic Review, Classification, and Analysis of the Current State-of-the-Art

Abstract: In recent years, huge increase in attacks and data breaches is noticed. Most of the attacks are performed and focused on the vulnerabilities related to web applications. Hence, nowadays the mitigation of application vulnerabilities is an ignited research area. Thus, due to the potential high severity impacts of web application, many different approaches have been proposed in the past decades to mitigate the damages of application vulnerabilities. Static and dynamic analysis are the two main techniques used. In this paper, a new classification for web application input validation vulnerabilities is proffered. In addition, various techniques/tools that are used to detect them are analyzed and evaluated to apprehend their strengths and weaknesses. Thus, this paper provides both technical as well as literature countermeasures to input validation vulnerabilities. Moreover, various statistical distributions of the reviewed techniques were manifested and scrutinize in different aspects to reveal the perception of the prevailing techniques and the gaps in the literature. In addition, the most widespread metrics are also propounded.

SQLi Detection with ML: A data-source perspective

Abstract: Almost 50 years after the invention of SQL, injection attacks are still top-tier vulnerabilities of today's ICT systems. Consequently, SQLi detection is still an active area of research, where the most recent works incorporate machine learning techniques into the proposed solutions. In this work, we highlight the shortcomings of the previous ML-based results focusing on four aspects: the evaluation methods, the optimization of the model parameters, the distribution of utilized datasets, and the feature selection. Since no single work explored all of these aspects satisfactorily, we fill this gap and provide an in-depth and comprehensive empirical analysis. Moreover, we cross-validate the trained models by using data from other distributions. This aspect of ML models (trained for SQLi detection) was never studied. Yet, the sensitivity of the model's performance to this is crucial for any real-life deployment. Finally, we validate our findings on a real-world industrial SQLi dataset.