Recommender Systems (RS) have been popular in assisting users with their choices, thus enhancing their engagement with online services. News RS are aimed to personalize users experiences and help them discover relevant articles from a large and dynamic search space. Therefore, it is a challenging scenario for recommendations. Large publishers release hundreds of news daily, implying that they must deal with fast-growing numbers of items that get quickly outdated. News readers exhibit more unstable consumption behavior than users in other domains. External events, like breaking news, affect readers interests. In addition, the news domain experiences extreme levels of sparsity, as most users are anonymous.In this book, we provide a comprehensive introduction about Deep Learning architectures for RS and an effective neural meta-architecture is proposed: the CHAMELEON. Experiments performed with two large datasets have shown the effectiveness of the CHAMELEON for news recommendation on many quality factors such as accuracy, item coverage, novelty, and reduced item cold-start problem, when compared to other traditional and state-of-the-art session-based recommendation algorithms.
A model plant, Arabidopsis thaliana, duplicates its chromosomes without undergoing cellular division, in a process known as endoreplication. The primary objective of this study was to identify genes and proteins that specifically accumulate in endoreplicated nuclei in Arabidopsis thaliana. The secondary objective was to identify the biological function of unique domains found in Arabidopsis topoisomerase VI subunit B (AtTopVIB) that contributes to endoreplication. Using the AtTopVIB amino acid sequence and protein database search engine, I identified two unique domains to which I designated the insertion of the N-terminal domain (IND) and the extension in the C-terminal domain (ECD). These domains are well conserved between Arabidopsis and Oryza sativa (rice) but very unique in the entire family. I analyzed the localization of AtTopVIB in Arabidopsis protoplasts with yellow-fluorescent protein (YFP) in Arabidopsis protoplasts using split-luciferase.
Fractal image encoding is one of the famous lossy encoding techniques ascertain high compression ratio, higher PSNR and good quality of encoded image. It is the approach that uses self similarity property in natural image.The main drawback of fractal image encoding is time consumption in search of appropriate domain for each range of image blocks.There have been various researches carried out to overcome the limitation of fractal encoding and to speed up the encoder.The neighborhood search strategy used in spatial domain reduces the encoding time from linear time to logarithmic time.The proposed algorithm used k-nearest neighbor search to find the similar edge shaped range and domain blocks to be mapped on the basis of DCT lowest coefficients in horizontal and vertical directions and then using similarity measure, the virtual codebook was prepared to encode the image and then decoded the image using averaging of pixels of selected domains. The proposed approach explained in this book is compared with the existing method and this work gave high PSNR value, high compression ratio and reduced MSE computations with little decay in image quality which is acceptable.
Bond-Graphs based system representation and genetic programming based search and synthesis can be combined for automated design of mechatronic or multi domain dynamic systems. This design scheme can be turned into an intelligent automated design system by introducing a dynamic database or knowledge library of relevant information (gathered through multiple runs of the same design scheme), which tends to help the designer in initial and final stages of the design process. The strength of Bond-Graphs in handling variety and dynamic behavior of systems from multi energy domains coupled with evolutionary search approach for machine intelligence implemented through genetic programming, leads to an ideal platform for testing and enhancing the state of available knowledge and technology for evolutionary design of physical systems as complex as found in nature. This book offers an introduction to this research area with a brief discussion about certain aspects and verification of one proposed automated design methodology for evolving mechatronic systems designs using Bond-Graphs and genetic programming.
In this Book, a domain specific search engine was designed to increase the relevance of search results. In addition, a keyword suggestion system was developed to allow searchers to narrow down the scope of their search to topics within the domain being targeted.Indicating that, the domain specific keyword suggestion system developed is superior to the keyword suggestion system used in standard search engines.To further investigate the keyword suggestion domain, two different algorithms were used.The first uses keyword frequency to extract relationships between keywords while the second uses a simplified method of association rules to mine relevant keyword suggestions. In conclusion, it was observed that domain specific search engines and keyword suggestion systems can have a significant effect on the quality of the search experience of users. The results obtained can be generalized to other domains with some minor modifications, specifically changing the terms in the dictionary table used and using a collection of documents in the relevant domain
Research into drug using behaviour has often focused on developing theories to explain why individuals use drugs, yet directly asking an individual why they participate in this behaviour does not necessarily uncover their motivations, or usefully inform strategies which may lead to behavioural change. Khantzian s theory of self-medication is used to help structure a cross-disciplinary literature search across biopsychosocial domains and also the spiritual domain, for which there is a paucity of empirical research. 12 in-depth case studies are informed by the Biographical Narrative Interpretive Method. There is twin-track analysis of the data employing BNIM panel analysis and Grounded Theory techniques. Findings suggest drugs are used to achieve homeostatic balance in the biopsychosocial domains. Experiences of emptiness also arose in the data. Exploration of these revealed two distinct types of emptiness: deficient emptiness and a perceived spiritual emptiness. Evidence suggests that drug use can be considered as an attempt to self-medicate against the state of deficient emptiness in order to achieve a spiritual homeostasis .
Revision with unchanged content. We investigate the problem of descriptive learning--learning rules that describe the underlying structure of a domain--in rich, qualitative worlds. Previous approaches to this problem have searched for laws in top-down, enumerative fashion. We present algorithms that belong to an alternative, data-driven search paradigm. In our algorithms, search is guided not by relationships between the forms of the hypothesized rules, but by correlations in the data they represent. We exploit anomalies in this data, hypothesizing that that patterns that are unlikely to have arisen by chance must represent features of the domain. We describe data-driven methods that discover rules in both propositional and relational domains. We apply our methods to the problem of finding planning invariants: formulae that are true in every reachable state of a planning world. Our methods provide a novel inductive approach to this problem. They find invariants from just a few reachable-state descriptions. They discover laws comparable in quality and complexity to those discovered by specialized planning-invariant discovery systems that require a far greater deal of specialized knowledge about the domain.
The study was conducted to develop Lactoferrin and TonB dependent receptor database, prediction of lactoferrin like proteins in plant and other resources, feasible prediction of Lactoferrin like sequences including domain and motif of theses sequences and ancestral relationship of these proteins with lactoferrin using phylogenetic analysis. Database was developed after retrieving data from different servers by using similarity search by blastp. The MSAccess was used to developed backhand whereas front hand created by JSP, JAVA, HTML and CSS. Tomcat Server v5.0 used for connection between front-hand with back hand. Feasible structures was found by using blastp against protein databank as the lactoferrin like protein seq from Ricinus communis, found homology with 1N76 based on e-value and bit score with percent identities. This structure was shared common domains with regular expression of Homo sapiens lactoferrin.
Conventional kinesins are molecular motors that walk along microtubules to carry out a large variety of cellular transport processes. The conventional kinesin of the filamentous fungus Neurospora crassa is 3-4 times faster than its animal relatives. Its neck domain displays characteristic properties that is also reflected in a specific sequence pattern. In the present work the functional roles of the specific neck and hinge domains for mechano-chemistry, dimerization and regulation of fast fungal kinesins are investigated. To address this issue, mutant protein constructs were generated, and characterized in terms of oligomerization, motility and kinetic properties. The results show that the specific neck and hinge regions are not critical for the basic mechanism of fast fungal kinesins. However, they play an important role for the regulation of the motor in a dimerization-based mechanism. In addition, the specific neck sequences may have evolved to fine-tune kinesin's processivity by combining tight structural connection of the head domains with a flexible tether that facilitates the diffusive search for the next microtubule binding site.