Bøger udgivet af Mohd Abdul Hafi

Cancer continues to be a human affliction and is brought on by aberrant cells that divide out of control. Cancers get their name from the word crab (a Latin word); as a result, they attach to every region of the body obstinately, just like crabs do. Thus, the malignant growth has the potential to penetrate and damage neighboring structures, and spread to faraway locations, ultimately leading to death. Cancer, in both sexes and at all ages, can affect all living cells within the body. Multiple factors are involved in causing the disease [1]. Cancer is a global burden that is projected to rise as a result of human proliferation and age, as well as the adoption of cancer-causing lifestyles. Malignancies with a historically low prevalence, such as female breast, colorectal, and lung cancers, are becoming more common in economically developing nations [2]. The type of cancer that has been observed with prominent occurrence and fatality in men is lung cancer, which is preceded by prostate and colorectal cancer (on the basis of prevalence) and stomach and liver cancer (on the basis of fatality rate). Contrarily, breast cancer (BC) has been reported with the most occurrence and is responsible for causing mortality in females, trailed by colorectal and lung cancers (incidence wise); cervical cancer ranks 4th (mortality and incidence wise) [3]. Cancer is also accountable for 1 out of 3 premature fatalities attributable to non-communicable diseases worldwide and it is expected that this count will increase every year [4]. Lung cancer is a very prevalent kind of malignancy and a significant reason of cancer mortality in males. The preponderance of women's cancer cases is still diagnosed with BC and contributes as a factor to female deaths. Nearly 2.1 million cases of breast malignancy in women are identified globally in the year 2018, representing one out of four cancer cases in females [2].

Chronic enteric infection, which causes diarrhea and gastroenteritis, become a growing issue worldwide, including a large death as well as the burden of disease rate, especially in children in underdeveloped nations. Diarrhea is the foremost largest killer in newborns and adolescents globally, responsible for 25 Lakhs fatalities per year (Lomazzi et al., 2014). The death of around 5 Lakhs kids below 5 years of age each year as of serious, dehydrating Diarrhoea and dysentery occurs each year, with an additional millions of hospitalizations. These fatalities are most prevalent in developing countries with limited health care resources (GBD Diarrhoeal Diseases Collaborators, 2017). In South Asia and Sub-Saharan Africa, diarrhea accounts for about half of all fatalities of children under the age of five (Alebel et al., 2018). Over 13 Lakhs public has been infected by Campylobacter yearly (CDC, 2017). There is now emerging evidence to kids is more sensitive toward extensive behavioral and emotional fitness problems as a result of recurrent infections in their early childhood (Nataro and Guerrant, 2017). Diarrhoeal diseases are the most prevalent infections caused by contaminated food, with 550 million people becoming affected each year, including 220 million children under the age of five, according to the WHO survey (WHO 2018) Every year, 3-5 billion individuals around the world are affected by acute gastroenteritis. Infectious gastrointestinal infections can be transferred through contaminated food or drink, direct contact between people, or fecal-oral contact (Fletcher et al, 2013)

In India, more than 50,000 small, medium and large electroplating industries are present which are mostly scattered in urban areas and discharge various heavy metals with their wastewater. Electroplating industrial wastewater has higher concentration of various heavy metal ions such as chromium, nickel, lead, copper, zinc etc. Small and medium scale industries mainly discharge their wastewater into municipal sewers. Ni, Zn and Cr metals are mostly found in electroplating industrial wastewater at high concentration. Chromium is extensively used in industries, such as chemicals and pigments, textile dying, tanning industries, wood preservation and finishing of plastics, metals and leather as well as in electroplating for surface treatment etc. In nature, cationic Cr(III) and anionic Cr(VI) forms are the most stable and dominant oxidation states of chromium (Miyata et al., 2007). Chromium (VI) is more toxic than Cr(III) (Sarkhosh et al., 2016) and it is a carcinogenic and mutagenic chemical (Enterline, 1974; Singh et al., 2011; Thatheyus and Ramya, 2016). Therefore it is extremely crucial to take care of the industrial wastewater having Cr(VI) before its release into the environment. The biogenic manganese oxides have huge potential for oxidation of Cr(III) to Cr(VI). Cr(VI) is typically movable, soluble and bioaccessible form and successfully sorbed on the surfaces of these biogenic oxides.

Diabetes mellitus is a chronic metabolic disorder caused by partial or complete insulin deficiency, resulting in hyperglycemia leading to acute and chronic complications. This disease is characterized by increase in the blood glucose level. This can mainly be attributed to the lack of insulin production (WHO, 2018). Diabetes mellitus is classified in to three types, Type 1, Type 2 and Type 3. Type-I diabetes mellitus (T1DM) also known as insulin dependent diabetes mellitus (IDDM) and results from failure of the beta cells of the pancreas to produce sufficient insulin. Inability to produced sufficient insulin can be traced to the body attacking itself or immune system attacking cells of the pancreas (Meglio et al., 2018; Chaytor et al., 2019). Type - II diabetes mellitus (T2DM) also known as non-insulin dependent diabetes mellitus (NIDDM) and commonly affects adults. It is the failure of the cells of the body to respond to insulin. In this case, insulin is produced adequately but there is an insulin resistance mechanism in action. As the disease progresses, the beta cells of the pancreas may fail or stop to produce insulin. Common cause of this type of diabetes is excessive body weight or obesity (Zheng et al., 2018; Paramithiotis et al., 2019). Type - III Gestational Diabetes occurs only during pregnancy. Treatment involves controlling diet, mostly sugar intake. Gestational diabetes might some times cause complications to the faetus such as death of child during delivery or shortly after delivery. Gestational diabetes also increases the risk of the mother developing other types of diabetes later in life (Hedderson et al., 2018; Jones et al., 2018).

In a network, usage of internet plays a crucial role. Transmissions of Data over networks are increased day by day. So, there is challenge for Future Networks regarding to data awareness. International Telecommunication Union -T recommendation (ITU- T) suggests a new network approach" Data Aware Networking" (DAN) [12]. Data Aware Networking is a revolutionary technology in the area of data maintenance and usage. In a distributed environment DAN has the capability to handle voluminous of data. If any user in network requests for Data Object, DAN provides the same Data Object as per user requirement. If users in the network increase automatically there is an issue called "flash crowd". Caching is a design goal of DAN to overcome the issue "flash crowd" [11]. Contextual Caching System provides the Data Objects either using Reactive Caching method or Proactive Caching method as per end user's requirement.

Cryptology is the branch of science that deals with the safe passage of information between the end users ensuring protection of the same from prying entities. Such as in the case of World War II when the need for secrecy peaked, the world witnessed evolution of efficient encryption techniques. Thus the information technology era came to pass. A common public channel, being vulnerable to breaches of secrecy being the channel of transmission, the need to secure information becomes undeniable. Images have encased lines- long ambiguity into a handful of pixels, making them an effective candidate for information transfer. Images have become an inevitable member in a plethora of firms, right from centers of governance to the panchayat hospitals resting in a mundane village. Hence the pressing need for an efficient and fool-proof image encryption system is justified.

Data is growing in the exponential order in last 3-4 decade; hence, the knowledge extraction and analysis is becoming difficult. However there exists various data mining algorithm and automatic tools for this purpose, one opposite end of this scenario is that, where data is not available in the sufficient volume we are unable to extract the useful knowledge. In this case, we cannot achieve the quality information or knowledge from the data. To resolve such situation an efficient algorithm for data farming is required. Data farming is the process to grow the datasets, similar to growing crops in agriculture. Data farming steps are data fertilization, data cultivation, data plantation & data harvesting. Data farming process is described in detail in chapter 3. This thesis presents a research work on the data farming algorithms. In this thesis, we proposed data farming algorithms for cardiac patient's dataset which includes temporal impact of the events like (1) diabetic, (2) myocardial infarction (MI) or heart attack (3) revascularization by percutaneous transluminal coronary angioplasty PTCA and (4) coronary artery bypass grafting surgery CABG etc. Proposed algorithms are also useful to predict the trends of future 'dose' required to the patients and provide the guidelines to the patients for precaution. So that, the amount of "dose" of the medicine either increases or decreases for the cardiac patient. Temporal aspects are rarely available in the literature associated with data farming methods and algorithms.

The fluid motion past a stretching cylinder has significant applications in the industrial manufacturing processes such as synthetic filaments, rubber, glass, fiberglass production, fabrication of papers, annealing and tinning of copper cords. On the other hand, many physical and manufacture coating processes have the boundaries in cylindrical configuration. Due to this, the influence of heat transfer past a cylindrical extended surfaces are critical in many mechanisms such as fiber spinning, hot rolling, and wire drawing. In view of this, many scientists worked on the fluid flow along a stretching cylinder. For instance, Wang [1] studied the viscous fluid flow through a stretching cylinder, and provided the exact similarity solutions. Batool et al. [2] used homotopy analysis technique (HAT) to explain the transport of viscous fluid over a stretching cylinder. Mukhopadhyay and Ishak [3] analysed the axisymmetric convective flow of Newtonian fluid towards the stretching cylinder using well-known shooting technique. Fang et al. [4] discussed the motion of unsteady viscous fluid towards a stretching cylinder using a numerical scheme. Munawar et al. [5] have investigated the unsteady boundary layer laminar motion of Newtonian fluid through analytically. Ali et al. [6] have presented the numerical analysis for the viscous flow towards a porous cylinder with variable radius. Abbasi et al. [7] studied the stagnation point propulsion of viscoelastic liquid using finite difference method. Vajravelu et al. [8] analysed the features of thermophysical properties on the transport of Newtonian fluid towards the cylinder using Keller-box numerical method. Ashorynejad et al. [9] studied the propulsion of nanofluid due to a stretching cylinder by employing of Runge- Kutta scheme (RKS). The viscoelastic fluid flow over a stretching cylinder using HAT has been explored by Bilal [1O].

Neurodegeneration is a serious, significant and major health threat worldwide. It is classified as a growing defacement of cognitive or motor function, which is followed by various types neurological disorders such like Alzheimer's disease (AD), Parkinson's disease (PD), Huntington's disease (HD), and amyotrophic lateral sclerosis (ALS) and many others with the no proper and standard treatment available so far Cummings J; Pihlstrøm et al. [1,2]. AD is the most leading cause of death in United States with ranked sixth, while PD the second most common neurodegenerative disease (NDDs) after AD, damage 1 to 2% person aged above 65 Silva et al; Rijk et al; Bekris et al; Farrer Alzheimer's Association [3-7]. The pathophysiology of these diseases is very diverse in nature Abeliovich et al; Canter et al [8, 9]. Some occurs due to decline in memory and cognitive function, other result in affecting the person ability of speaking. walking. breathing Taylor et al: T.Wyss-Corey [10.11]. The most of the neurodegenerative disease are describe by their specific protein aggregation and anatomic predisposition. Various underlying processes associated with increasing neuronal cell loss and death includes proteotoxic stress, oxidative stress, abnormal ubiquitin-proteasomal as well as autophagy lysomal system, programmed cell death, mitochondrial dysfunction, neuroinflammation Dugger et al: Milenkovic [12, 13]. Various studies support the complex etiology of these neurological disorder such like abnormal or defective protein aggregation and degradation, reactive oxygen species (ROS) overproduction and oxidative stress, mitochondrial dysfunction and exposure to chemical/environmental toxicants such as metal ion, pesticides, pollution etc. Each neurological disorder has their own methodology and manifestation at molecular as well as clinical state. However some common pathways are identified in different disease pathogenesis including oxidative stress, mitochondrial dysfunction, protein misfolding. formation of free radicles, metal dyshomeostasis to study the underlying mechanism of neurodegenerative disease and their common pathogenesis among others. Exploring these similarities will open a new ray of hope for therapeutic invention that will provide better understanding of disease pathogenesis and improved treatment for neurodegeneration.

Cardio-vascular disease (CVD) is a type of abnormality of faithful heart function. These medical conditions relate to the abnormal health conditions that directly affecting the heart and all its components. Due to the negligently treating, every year, lot of people were maximally suffering by CVDs which is reported by The World health organization [1]-[4]. Among the CVDs, heart failure is caused by the heart inability in pumping required blood for the need of body and sudden heart attacks occur when normal blood circulation ceases because the heart fails contract effectively. Today's biotelemetry is an efficient tool to diagnose about the cardiac abnormalities if the patient is too far to take help from specialist [5]-[6]. The cardiac signal, also called the electrocardiogram (ECG) signal is extracted by a wearable acquisition device and sent through a biotelemetry system to the hospital. The doctor then analyses the signal to take decision for the appropriate action, and the decision is sent to the side where the patient is required for the quick action [7]-[8]. The main considering point here is the ECG signal was interfered with different types of artefacts during the acquisition. Among the artefacts, Power Line Interference (PLI), Baseline Wander (BW), Muscle Artefacts (MA) and Electrode Motion artefacts (EM) are interested as the commonly occurred artefacts. The ECG signal quality is very important in clinical monitoring and diagnosis. These artefacts affect the ECG signal quality. Hence, the need of investigation is high on this issue to extract the high resolution ECG signals from impure recordings which were contaminated with background noise. Therefore in interpretation of easy and accurate ECG, Signal enhancement by filtering is desirable to separate the valid signal components from undesirable artefacts.

In recent years deep structured learning is practiced in many safety-critical applications. Deep learning algorithms are used in every field, such as face recognition (laptop, smartphones) and fingerprint recognition for financial applications. A deep neural network performs classification tasks. In designing any application using machine learning, enormous training data are required. Appropriate training gives the correct results. Other than training, many factors need to be considered viz labelling data: (manual task), expert advice, model selection, etc. In 2013 researchers observed that deep neural networks give expected results to the adversary by using adversarial samples during the testing or deployment phase without directly accessing the model, considering no knowledge about the model.Adversary does use labels during the deployment phase, modify labels, and apply them to attack the victim model. Trustability of the deep neural network is a significant concern in many safety-critical applications, where input data contains noise and is not certain. Most frequently, uncertainty and noise in data are random, and adversary has incentives to change the data. As a preventive measure to detect malicious activities increases, adversaries become motivated to put extra effort into deceiving the algorithms. The security issue is then critical, especially for the applications which involve safety and monetary usage.

Quest for innovative ways for the promotion of inclusive growth has been a subject of vibrant academic debate for last many years. The 'easy access to credit' is identified as one of the important factors that enable people to avail growth opportunities. In the contemporary scenario, the financial institutions are becoming more risk averse and in their efforts to minimize non-performing assets, they are reluctant to lend to the people who don't have any stable source of income. At the same time, boosting investment rates in more decentralised micro, small and medium enterprises is also the major emphasis in contemporary developing economies. It is in this context, Islamic microfinance system finds its significance and offers the component for a paradigm shift in the development thinking. While conventional microfinance has been widely acclaimed for facilitating the poor to access financial services enabling their economic empowerment as well as freedom from the exploitations by the unauthorised money lenders, yet it has not adequately addressed their problems. Group-based Grameen model, individual lending, social business, credit union and village lending are the diverse models of microfinance existing in different parts of the world.

Cryosurgery (cryoablation or cryotherapy) is a surgical technique in which localised ap- plication of freezing temperatures is used to ablate benign and malignant tumours [1, 2]. It was the oldest method to treat skin cancer in the 18th century [3]. There are various conventional techniques available to treat skin tumours such as surgery, chemotherapy, moh's surgery and radiation therapy [4]. Cryoablation is applied to different types of solid tumours such as prostate, liver, breast, kidney and pancreas [5-8]. This method has cer- tain advantages over other conventional methods and also posseses a few limitations which have to be overcome so that it can be used in regular clinical practise by the sur- geons. Thus, this chapter presents insight towards the historical and modern techniques related to cryoablation, skin cancer, the suitable method to treat skin tumours, i.e, cryo- gen spray, imaging technique to determine the evolution of the ice front, and numerical studies to solve the bio heat equation.

Both theoretical and empirical research on workplace diversity have increasingly found place in organizational studies. Workplace diversity is defined as real or perceived difference between members of a group on any characteristics that may induce the perception that the other individual is different from the self (van Knippenberg & Schippers, 2007; Williams & O'Reilly, 1998). Most of the early research on diversity has focused on the demographic attributes of age, gender, ethnicity, race, tenure and functional background (Milliken & Martins, 1996; Williams & O'Reilly, 1998). In principle, there are infinite such attributes which might potentially evoke the feeling of being different, like education, experience, religion, value, belief, personality and so on. The readily visible and observable attributes of difference are referred as surface-level diversity attributes. On the other hand, few differences that are not so readily observable and are only experienced and felt by the individual when they are continuously interacting with group members. Such differences like for example, differences of value, belief are referred to as deep-level diversity attributes. There are two opposite lines of thought available in popular literature in reference to effects of diversity on individual or team outcome. The negative diversity outcome is associated with social categorization perspective in which an individual considers other group members as either in-group members or out-group members.

Metal Organic Frameworks (MOFs) are referred to as coordination polymers. These are a category of crystalline materials which are presently attracting a superb amount of attention because of their attractive properties [1]. Omar Yaghi et al., invented the word MOF in 1995. MOFs consist of organic and inorganic construction systems where organic ligands, (also called spacers) are combined with clusters of metal ions (also known as nodes) to generate expanded frameworks [2]. In some cases, the frameworks are sufficiently rigid to form internal voids after removal of the solvent, forming structures with high porosity and surface area (up to about 7000 m2g-1) (Figure 1.1) [3]. MOFs have incontestable prospective applications in an exceedingly broad spectrum of areas, as well as gas separation and storage, catalysis, sensing and drug delivery [4-7]. Metal ions that operate as connectors and organic bridging ligands (as linkers) square measure the cornerstone of the compound. The formation

In recent years, wireless sensor networks (WSNs) have developed as a potentially useful technology for a wide variety of purposes, from surveillance to environmental monitoring. Wireless sensor networks are networks of tiny, low-power sensor nodes that can collect, analyze, and transmit data wirelessly. To gather information about their environments, these nodes generally have sensors like temperature, humidity, and motion detectors. The limited energy and computing resources of the sensor nodes are a significant barrier to WSNs. Since sensor nodes have limited energy capabilities due to their tiny size and dependency on batteries, energy efficiency is a crucial factor in WSN design. In order to keep WSNs up and running for as long as possible, it is essential that their lifespan be maximized while their energy consumption is kept to a minimum (Akyildiz et al., 2002). Many other energy-saving methods and procedures for WSNs have been offered by scientists as potential solutions to these problems. The goal of these methods is to reduce power consumption across the board, from individual sensor nodes to the network as a whole. In order to decrease the amount of energy needed to transmit data between sensor nodes, many clustering algorithms have been developed. Clustering is often employed in WSNs to increase scalability, decrease energy consumption, and consolidate data (Heinzelman et al., 2000). Clustering is a method for grouping sensor nodes, with the cluster head (CH) of each cluster being in charge of coordinating the work of the other nodes in the cluster (Heinzelman et al., 2002). The cluster heads gather, analyze, and send data from their clusters to the base station (BS), which helps save power and extend the life of the network.

The fast research development of electronic devices, optical devices, micrometer and nanometer sized electronic and optoelectronic devices are in very high demand because of their high performance, fast functionality and very small power consumption. Nanoscale devices had replaced so many complicated and large sized devices. There are so many applications of nanodevices and their use is increasing day by day. Based on recent discoveries in nanodevices, Photonics and plasmonics come into their existence in the last 10 years. For past 40-50 years only some of their basic concepts have been known. Photonic materials and devices have played an invasive role in communications, energy conversion, and sensing since 1970s. In Broad sense, over the next 10 years nanophotonic structures and devices promise drastic reductions in energies of device operation, densely integrated systems with lower power dissipation, enhanced spatial resolution for imaging and patterning, and new sensors of increased sensitivity and specificity. (Roco, Mirkin and Hersam, 2011) The branch of science that deals with the behaviour and properties of light is known as optics. Optics generally depicts the behaviour of visible, ultraviolet and infrared light. Optics is the branch of physics optics in which focuses on manipulation of light. Mostly all the optical phenomenon uses the classical electromagnetic description of light. Light has both wave like as well as particle like properties which can be explained with the help of quantum mechanics. Optics has found its various applications in the areas of Astronomy, Engineering, photography and medicines.

It is observed globally that the number of Diabetes cases are increasing in every country which would have serious consequences in the society, if it is not properly checked and handled by the researchers and physicians it would be too late for the health benefit of mankind. India is a country which has been endorsed by WHO that it is going to become second largest affected country in the world. So, pharmaceuticals companies in India and abroad are carrying out researches in search of allopathic and ayurvedic medicines for the purpose of treating and curing diabetes patients. By definition diabetes is an enduring condition caused by inadequate amount of release of insulin or inadequate production of insulin in the body or due to obstruction created in the path of insulin secretion that results in lessening amount of carbohydrate, fat, and protein metabolism. Medical literature on diabetes has categorized two types of diabetes that is, Type -1 or insulin dependent; in which an individual is totally depend on insulin which is given to a patient through external sources, whereas Type-2 or non-insulin dependent; in which an individual is treated through medicines or in severe cases they are also given insulin along with medicines.

Blood coagulation is a highly conserved and spontaneous physiological response after vessel injury. It restricts the excessive blood loss from our body. It includes a plethora of plasma reactions resulting in the formation of thrombin and a series of tightly regulated intracellular reactions mediated by platelet activation. In the last century, our understanding of the mechanisms and effects of blood coagulation had made significant development. Beginning from the early clinical observations of rare patients with genetic clotting disorders, our vision has reached to the introduction of newer technologies: from blood coagulation experiments to protein chemistry, to gene targeting technologies and to cure hereditary disorders by external medicinal assistance. Recently in the last decade, several studies have suggested an important contribution of blood coagulation factors in tumor progression. Here in this thesis, I have tried to explore some unanswered questions regarding i) The activation of cancer signaling pathways by clotting factors in the tumor microenvironment and its impact on primary tumor growth. ii) Intravascular activation of blood coagulation and its impact on tumor metastasis. iii) Targeting the molecular mechanism of the inflammatory ability of cancer cells to improve the therapeutic approach against cancer progression.

In recent years, MEMS and NEMS technologies are the new rising technologies for an assortment of applications because they posses many advantages such as low power consumption, reliability and low mass etc. The both hard and soft magnetic material which includes plays a significant role for the manufacturing of MEMS devices. In particular soft magnetic materials are the appropriate one for making MEMS devices because of their outstanding magnetic properties. Since the exploration basically centers around the magnetic properties of attractive NiFeP materials, let us discuss the fundamental ideas of magnetic materials in this chapter. The materials which can be magnetized by the application of external magnetic field are known as magnetic materials. Magnetic materials are commonly classified into five categories in terms of magnetic properties of the atomic dipoles and the interactions between them. 1. Diamagnetism 2. Para magnetism 3. Ferromagnetism 4. Antiferro magnetism 5. Ferri magnetism

Cloud computing, in general, is the delivery of hosted services over the internet. Private clouds use a company's own infrastructure, whereas public clouds use shared infrastructure. Since they are by definition hosted by a public cloud provider like AWS, Azure, or Google Cloud, public clouds reduce the burden of managing the infrastructure. Enterprise data and application code are stored on the servers of the cloud service provider in a public cloud deployment. On the other hand, private clouds are hosted on a business's internal network and are frequently firewalled and physically secured. Largely on-premises infrastructure-invested mature corporations frequently use that capital to build their private clouds. Other names for cloud computing include utility computing and on-demand computing. Cloud services fall under three major types or categories of cloud computing: Platform as a Service (PaaS), Infrastructure as a Service (IaaS) and Software as a Service (SaaS). The dynamic nature of infrastructure management, especially when increasing applications and services, can offer a variety of problems even when businesses adequately resource their departments. They can outsource many of the time-consuming IT-related duties as-a-service models. Understanding the security needs for data protection is crucial as organizations continue to shift to the cloud. The management of this infrastructure may be transferred to a third-party cloud service provider, but the accountability and security of data assets may not always follow.

The world of twenty-¿rst century runs using computer programs. Whether it is a hand- held gadget or a powerful supercomputer, humans write programs, often in a high-level programming language (such as C, Java, Python, R, OCaml, and so on), to commu- nicate with these digital devices. However, in order to execute a program on a given machine, the program ¿rst has to be converted to a form that can be understood by that machine. Akin to the communication between any two persons who speak and under- stand different languages, human-computer interaction, thus, also requires a translator. Two primary examples of translators that convert the high-level language of computer programs to the low-level language of machines are interpreters and compilers. An interpreter executes a program instruction-by-instruction. On the other hand, typical compilers, while translating code from one form to another, perform a series of analyses on the input program. The results of these program analyses are used to drive a plethora of applications: understanding the semantics of a program, debugging an erroneous program, ¿nding out whether a program is secure, and optimizing the program in terms of the resources it may consume (time, memory, energy, etc.), to name a few. Depending on the time at which a compiler translates a program, there are two kinds of compilers and program analyses: static and just-in-time (or JIT).

The realm of education is undergoing a profound transformation, catalyzed by the rapid advancements in technology and the growing needs of learners. In this dynamic landscape, Massive Open Online Courses (MOOCs) have emerged as a transformative force, redefining traditional educational paradigms and democratizing access to knowledge. MOOCs offer a promising avenue for learners across the globe to engage with high-quality educational content, transcending geographical and socio-economic barriers. As a result, they have garnered widespread attention for their potential to revolutionize the way education is delivered and consumed. However, alongside the remarkable promise of MOOCs, there also exists a critical challenge that has garnered attention from researchers, educators, and policy makers alike - the issue of high attrition rates within these online learning platforms. The allure of open access and flexibility that MOOCs offer is often accompanied by the concern of learners not completing the courses they enroll in. This challenge poses questions about the effectiveness of MOOCs in engaging learners and delivering meaningful learning expenences. This chapter serves as the gateway to an in-depth exploration of the multifaceted landscape of MOOCs and their challenges. It provides an overarching framework for understanding the research that follows, offering a comprehensive view of the issues, opportunities, and complexities inherent to the realm of online education and MOOCs. By navigating through the historical development of MOOCs, exploring the intricate factors contributing to attrition, and delving into the possibilities offered by artificial intelligence and machine learning, this chapter lays the foundation for unravelling the tapestry of MOOC attrition.

The dictionary meaning of the word 'restoration' is the act of bringing back a system, that existed previously. In data processing, 'restoration' refers to the task of recovering the original or clean data from the given corrupt data. In Computer Vision, restoration is considered to be an umbrella term as it spans over a variety of problems that depends upon the type of: • Data being restored. • Corruption, the data has to recover from. The most frequently dealt data in Computer Vision problems is either an image or a video, and the most typical form of corruption, found in this data is due to Blur, Noise, Haze, Rain, Missing-pixels or Low-resolution, among others. The restoration problems with the stated corruptions can be but not limited to Deblurring, Denoising, Dehazing, Deraining, Inpainting and Super-resolution. These can be applied either on an image or a video. The next section elaborates some of these problems that will be addressed in this thesis.

In manufacturing of components, use of fixture is unavoidable. The proper fixture design is required to maintain the form and dimensional accuracy of the component under machining operations. In machining of the components, the fixtures are used to locate and constrain the workpiece to perform the machining effectively. The use of fixtures ensures the accuracy, reliability and interchangeability of the components. It enhances the economy of manufacturing by reducing setup time and can be handled by less skilled manpower. The effective machining is achieved by Workpiece Fixture System (WFS) in a batch or mass production system. The workpiece is located and constrained in WFS using number of locators and clamps. The arrangement of locators and clamps around the workpiece is termed as Fixture Layout Design (FLD). The effective FLD ensures the quality of the components by minimizing the elastic deformation of the workpiece under machining conditions. The key elements of fixture system used for any machining operations are Fixture body, locators and clamps. Fixture body acts as the base element of the fixture system on which the component is placed for performing machining.

The focus of work in this thesis is on the problem of Distance Metric Learning (DML) with limited supervision, i.e., we seek to learn distance metrics with very few labeled examples (semi-supervised DML) or no labeled examples at all (unsupervised DML). The DML is fundamentally related to the problem of embedding learning, and hence both the problems are studied interchangeably. Though the recent, state-of-the-art DML approaches have shown promise of success on a variety of machine learning tasks, they all make use of class labels or manual annotations, to obtain constraints for DML. However, obtaining class labels for large datasets may not be feasible in many tasks, thus limiting the practical applications of these approaches. To further emphasize the importance of looking beyond labeled data, we begin the thesis by borrowing the analogy recently stated by Yann LeCun 1: "Assuming intelligence to be a cake, supervised learning can be seen as the icing and Reinforcement Learning (RL) can be seen as the cherry. It is unsupervised learning that forms the bulk of the cake!"

concerned with the study of recognition phenomena and studies all kinds of interactions where no formation of covalent bonds with reacting species. Lehn described supramolecular chemistry as "the chemistry beyond molecules" [Lehn, 1988] The importance of this field was emphasized further when Lehn, Cram and Pederson were awarded Noble Prize for their work on molecular recognition in 1987. The host-guest type complexes formation was a break through development where a guest specifically binds to a host. Cyclodextrins are more accurate and important cyclic structures, among various studied hosts. They have been described as "Unique seductive molecules" [Mandalia et al.,2012]. Cyclodextrins were discovered by Villiers in 1891, isolated CDs from a culture of Bacillus amylobacter, grown on a medium of starch. The crystalline product obtained was named as cellulosine by Villiers since it resembled cellulose [Villiers,1891] In 1903, Schardinger isolated and study CDs, on the basis of his studies CDs are also referred to as Schardinger's dextrins [Schardinger,1903]. In the early literature CDs were also referred as cycloamylose, cycloglycoamylose etc. y-CD (cyclomaltooctaose) was discovered by Freudenberg[ Frendenberg etal.,1935 ]. The structure of CDs established by Freudenberg, and issued by Ewald and others [Ewald et al.,1936, Karrer etal.,1921, Miekeley,1932]. In 1950s, French and co-workers observed the chemical process for the manufacturing of CDs and larger CDs alivement was also confirmed [French 1957].Firstly Cramer, uncover the potentialability of CDs as complex forming agents and assessed the ability of CDs to form a complex with different types of drug molecules [Cramer et. al, 1954

Sustainable energy supply has become the grand challenge of the twenty-first century from the political, economic and societal point of view (Zandler et al., 2016). The global energy sector (Electricity and transportation) has already caused serious environmental issues because of Greenhouse Gases (GHGs) emissions, which are emitted directly or indirectly to the atmosphere. Greenhouse Gases (GHGs) emissions lead to global warming, climate change, environmental pollution and limitations in the conservation of natural resources (Jibran and Mudassar, 2016; Kweku et al., 2017; Gwani and Abubakar, 2016). The utilisation of fossil fuels in energy-related issues is among the biggest culprit of anthropogenic carbon emission accounting for about 70% of the global CO2 emission. Electricity as the backbone of development of any society has already contributed 37.5% of the total CO2 emission in the globe, releasing 7700 million tonnes of CO2 annually. This is an indication that serious transition is necessary for decarbonising electricity across the globe (Moutinho and Robaina, 2016; Safari, 2011; Shata, 2012). Electricity is one of the most widely used forms of energy by humans throughout the world. However, literature scouring has revealed that more than 1.64 billion people in the globe, are living completely without access to electricity, of which the significant percentage coming from rural Africa and Asia (Samsul Alam, 2018; Evangelos et al., 2016). With this significant population living without electricity, one can conclude that, in some countries of the globe, connection to electricity is rather a privilege, not a right. Astoundingly, majority of the population living without electricity across the globe, are low-income houses in the villages where renewable energy resources are readily available and cheaper to explore, when compared with grid expansion (Sampo et al., 2011; Dihrab and Sopian, 2010; Finnveden, 2009).

The fundamental unit of heredity is DNA (Deoxyribonucleic acid), which carries the ge- netic information of living species. DNA is a complex molecule where each molecule consists of two strands twisted to form a double helical structure[11]. The backbone of strands is made up of alternative sugar and phosphate groups, and these strands are bonded together with base pairs formed from four bases: Adenine(A) with Thymine(T), and Cytosine(C) with Guanine(G). The bases are organized in a sequence one after an- other to form a genetic code. Decoding the code extracts the information contained inside the DNA! With this knowledge in mind Human Genome Project (HGP) [12] was initiated in 1990 with a cost of 2.7 billion dollars to read the human genome, which has approx 3 billion base pairs. The project was closed in 2003, reporting a 1.3 billion base-pairs count of DNA in the human genome. A lot of engineering techniques and technologies have been developed over the years to extract information about the organism. Generation after generation, various techniques have been introduced to reduce the time and cost of DNA sequencing.

The evolution of angiosperms made a paradigm shift in the biology and ecology of animals that lived at that time (Bascompte and Jordano 2007, Bronstein et al. 2006, Endress 2011, Regal 1977, Willmer 2011). The herbivory radiated to various levels, and new traits and taxa evolved once the flowers with pollen grains, nectar, and fleshy fruits evolved (Herre et al. 1996, Mitchell et al. 2009a, Ramos and Schiestl 2019, Strauss and Irwin 2004). In insects, pollen and nectar collecting behavioural and morphological traits and digestive system have been evolved. While many insects used these plant resources as alternate food, many lived only at the cost of pollen and nectar. As a result, plants started relying on animals for dispersing their pollen grains and seeds, and made them an inevitable and reliable mutualistic partner in their reproductive biology. Likewise, many animals have been evolved to use protein and carbohydrates of flower origin. Plant-pollinator and plant-frugivores interactions, thus grew-up as two strong mutualistic interactions between plants and animals. Man depends on these interactions, in particular the plant-pollinator interaction, extensively in their agricultural systems. Today, pollinators have a value of several trillions of rupees or dollars (Potts et al. 2016), particularly for the countries that have an agricultural economy (Gallai et al. 2009).