Research
Ongoing Research
1.Machine Learning on Geospatial Data
The way we handle spatial data has changed
significantly as a result of the integration of machine learning with
geospatial data. Machine Learning techniques increase the accuracy, efficacy,
and capacity to analyse vast amounts of geospatial data by using algorithms to
find complex relationships and patterns in spatial information. Machine
Learning helps decision-makers make well-informed choices by providing better
insight from data, predicting future trends, and making the greatest use of
available resources. In the current paper meteorite
landings data set is chosen. At first preprocessing of data set is demonstrated
followed by the data visualization using Basemap Python library, which is an extension of Matplotlib, makes it
easier to plot 2D data on maps. Regression is performed on the data set to
predict the year of meteorite landings. Python is used for implementation
because of it ease. Accuracy of the regression model used for predicting the
year of fall of meteorite is computed along with the regression line.
2.Exploring the Semantic Web in Healthcare
In today's world, huge
emphasis must be given to healthcare practices and developing good Health care
products to serve the ongoing need in the health sector. There are many challenges,
which arise due to the collection and proper sharing of medical data, which
consists of data from many different sources in varying levels of abstraction
and diverse formats. Semantic Web technologies are used to overcome these
challenges. The Semantic Web represents knowledge consistently by using ontology
so that the information from various medical repositories can easily be shared reliably.
Apart from just storing and sharing medical information, the Semantic Web also
provides an enormous potential for performing inferences on the stored
knowledge for performing novel discoveries and correlations in the medical
field. This paper provides an insight into using Semantic Web technology in the
healthcare domain by proposing SemHCare Architecture.
3.E-Learning on Semantic Web
E-Learning is the current
trend that is gaining more significance post-pandemic, E-Learning has several
distinctive features which has attracted many users across the globe and there
is a huge demand for E-Learning content. E-Learning is the delivery of learning
material by any electronic gadget, the communication channel used for the
delivery of learning material is the World Wide Web. Currently, the content
displayed on the World Wide Web is only understood by humans but not by
machines, in order to make machines understand the content and make intelligent
decisions the World Wide Web can be replaced by the Semantic Web. In the
current paper, SemELearn a Semantic Web-based E-Learning system is proposed
that utilizes the underlying characteristic feature of the Semantic Web.
4.Exploring Reasoning for Utilizing the Full Potential of Semantic Web
The Semantic Web can infer additional
information from existing information by using a reasoning mechanism, this
reasoning mechanism is not present in the current Web. With an increase in
Semantic Web data advanced reasoning methods must be formulated which must be
error-free and must ensure that no inconsistencies are generated within and
between data. Ontologies form the backbone of the semantic web. Although OWL
supports reasoning, they are incapable of providing the inference needed in the
Semantic Web. OWL is established on an open-world assumption that does not
support the required inference in the Semantic Web. This can be overcome by
integrating rules with OWL. Rules along with OWL provide the required inference
support in Semantic Web. A significant role is played by rules in Semantic Web
layered architecture. Rules offer the basics for performing automated
reasoning. Rules can be written to incorporate business logic and constraints in
Semantic Web applications. The design of rules for supporting the Semantic Web
was documented as a most important design issue by Sir Tim Berners Lee. Rules
are used to reason about the data. Various types of reasoning can be applied on
Semantic Web. Logic programming can be integrated with the Semantic Web concepts
to create advanced reasoning techniques. In this chapter significance of
reasoning in the Semantic Web is highlighted with an emphasis on reasoning
performed by rule engines. Apart from this, types of rules and rule languages
along with the set of reasoners used for performing inference are also
explained in detail.
5.Quantum Machine Learning with Super Computing Powers on Cloud Environment for Real Time Applications
As a result of quantum theory, quantum computing is the study of novel computer based technologies based on these ideas. The quantum nature and behaviour of energy and matter are described by quantum theory (at the atomic and subatomic level). In quantum computing, specific computational tasks are handled by combining different bits. All of these new technologies are also considerably more efficient than their traditional counterparts. The creation of these machines represents a significant leap forward in computing power due to the rise of quantum computers, which offer enormous increases in performance for specific applications. While this is true, quantum computing excels at tasks such as modelling networks, machine learning, and AI. IBM Qiskit is the main platform for implementing quantum machine learning, according to the paper, and the predictions made using PennyLane were more accurate. For real-world engineering issues, PennyLane supports the qubit and CV quantum computing models, both of which are based on the qubit model. In reality, a hybrid computer using both qubit and CV quantum nodes is possible.
6.Transforming Medical Data in to Ontologies for Improving Semantic Interoperability
Health Care Systems today has mostly become patient-centric and digitally expressed in form of Electronic Health Record (EHR). The medical data collected using clinical codes from multiple sources are saved in structured form or free text. This data heterogeneity has extensively increased due to the exponential growth of health care data which makes data extraction complex, creates interoperability issues, and hinders healthcare development. Web Ontology Language (OWL) combined with Semantic Web technologies adds simplicity to searching, integrating, reusing, sharing information, and addressing interoperability issues. It is very much essential to transform medical data into ontologies to achieve improved Semantic Interoperability. This chapter focuses on a review of the significance of the semantic web in the medical industry and discusses different methods of building or transforming open EHR-based medical data to OWL individuals. Finally, the chapter gives an insight into ontology mapping for semantic interoperability with its benefits and also considerable challenges.
7.Big Data Analytics for Big Medical Data-Tools Applications and Challenges
Internet has transformed the computer and
communications world which brings loads of information at our fingertips,
resulting in rise of Big Data at an unbelievable momentum. Every organization today performs different
analytics on the big data for making
better and potential decision making. Health organizations are also adding to
the increase of big medical data through sources including electronic medical
records of hospitals and patients, medical examination reports, and medical
device results and many more. Smart digital and wearable devices create vast amounts of heath data adding more to
medical big data. Effective analysis of this data is always essential to improve the health care system. The existing traditional data analytics tools
and frameworks may not be capable enough to
process this voluminous data . Today many systems are adopting Big
data tools and techniques for advanced data analytics and better
decision making. This paper introduces big medical data in healthcare ,
applications and benefits of Big Data analytics in healthcare. We also present
the technological progress of big data
and Challenges in healthcare
systems.
Completed Research
SemRPer- A Rule based personalization system for Semantic Web
With millions of pages available on Web, it has become difficult to access relevant information. One possible approach to solve this problem is Web personalization. Web personalization is defined as any action that customizes the information or services provided by a Web site to an individual. Personalization in Web is no longer considered an option but has become a necessity because of the movement from traditional physical stores of products or information to virtual stores of products and information. Personalization helps to solve the customer retention problem. Web contains documents which can be interpretable only by human beings but not by machine. To support machine-processed content on Web, Tim Berners-Lee proposed Semantic Web, which will enable the machines to understand and process the information automatically by adding meaning to the documents on Web. World Wide Web has been extended as Semantic Web for supporting extraction of new knowledge to facilitate decision-making processes by enabling machines to understand the content. When personalization is applied to the Semantic Web it offers many advantages when compared to the traditional Web because Semantic Web integrates semantics with the unstructured data on Web so that intelligent techniques can be applied to get more efficient results. Semantic Web is the next generation Web where information is organized into conceptual spaces called as ontologies. Ontological relationships are utilized in inferring additional characteristics required for producing personalized recommendation for Semantic Web users. SemRPer- A Rule based personalization system for Semantic Web users is proposed in this research work. Jena Semantic Web framework is used for creating Semantic Web application because it’s free; Java based, contains API and supports plug-ins and tools for working with RDF and OWL. SemRPer system generates personalized recommendations using a generic rule reasoner which can be configured using rule set which consist of a set of rules written to carry out the personalization task. Logic formalisms supported in Semantic Web is utilized for writing rules. SemRPer system performance is demonstrated on ontologies from different domains. Precision, Recall and F1-Measure measures are used to specify the quality of generated recommendations. SemRPer system generates recommendations with high values of precision, recall and F1- Measure when compared to recommendations generated by existing system. SemRPer system is rigorously tested based on the performance markers such as loading time of ontology, configuration and reasoning time of the reasoner and the memory consumed for generating personalized recommendations. Performance of the proposed system is found to be far ahead in terms of existing system. The main aim of personalization system is to reduce the time required for searching the required information is successfully achieved in this research work by generating a novel Semantic Web based personalization system which not only produces high quality recommendations but also has a very good performance in terms of time and memory consumed to generate recommendations.