Fine-tuned Approach Research Articles

Sliding-Window Pitch-Class Histograms as a Means of Modeling Musical Form

This study proposes a fine-tuned approach to modeling musical form in Classical repertoires by analyzing pitch-class distributions in symbolic data using machine learning algorithms. Results suggest that sliding-window histograms, which take the temporal component into account, palpably improve algorithms’ performance in assigning form labels to pieces. Pitch-class histograms were extracted from major-mode piano-sonata movements by W. A. Mozart and L. v. Beethoven according to two methods: whole-piece histograms and sliding-window histograms. In the latter method, richer features were obtained by calculating histograms for 9–90 partially overlapping windows per piece. Supervised learning methods, such as Support Vector Machines (SVMs) and Artificial Neural Networks (ANNs), yield a good separation of data points according to three distinct form labels: sonata form, ABA (ternary) form, and variation form. For both models, using sliding-window histograms significantly improves performance with regard to the whole-piece histogram data. Unsupervised clustering into three components employing a Gaussian Mixture Model (GMM), on the other hand, yields no successful results. Finally, we offer an in-depth exploration of our findings, and propose some directions for further research based on sliding-window data.

Deep transfer learning for alzheimer neurological disorder detection

Alzheimer’s disease is becoming common in the world with the time. It is an irreversible and progressive brain disorder that slowly destroys the memory and thinking skills and, eventually, the ability to perform the simplest tasks. It becomes severe before the noticeable symptoms appear and causes brain disorder which cannot be cured by any medicines and therapies, however its progression can be slow down through early diagnosis. In this paper, we employed different CNN based transfer learning methods for Alzheimer disease classification. We have applied different parameters, and achieved remarkable accuracy on benchmark ADNI dataset. We have tested 13 differnt flavours of different pre-trained CNN models using a fine-tuned approach of transfer learning across two different domain on ADNI dataset (94 AD, 138 MCI and 146 NC). Comparatively, DenseNet showed better performance by achieving a maximal average accuracy of % 99.05. Significant improvement in accuracy has been observed as compared to previously reported works in terms of specificity, sensitivity and accuracy. The source code of propose framework is publicly available.

Conjugation of Native-Like HIV-1 Envelope Trimers onto Liposomes Using EDC/Sulfo-NHS Chemistry: Requirements and Limitations.

The display of native-like human immunodeficiency virus type 1 envelope (HIV-1 Env) trimers on liposomes has gained wide attention over the last few years. Currently, available methods have enabled the preparation of Env-liposome conjugates of unprecedented quality. However, these protocols require the Env trimer to be tagged and/or to carry a specific functional group. For this reason, we have investigated N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide/N-Hydroxysulfosuccinimide (EDC/Sulfo-NHS) chemistry for its potential to covalently conjugate tag-free, non-functionalized native-like Env trimers onto the surface of carboxyl-functionalized liposomes. The preservation of the liposome’s physical integrity and the immunogen’s conformation required a fine-tuned two-step approach based on the controlled use of β-mercaptoethanol. The display of Env trimers was strictly limited to activated liposomes of positive charge, i.e., liposomes with a positive zeta potential that carry amine-reactive Sulfo-NHS esters on their surface. In agreement with that, conjugation was found to be highly ionic strength- and pH-dependent. Overall, we have identified electrostatic pre-concentration (i.e., close proximity between negatively charged Env trimers and positively charged liposomes established through electrostatic attraction) to be crucial for conjugation reactions to proceed. The present study highlights the requirements and limitations of potentially scalable EDC/Sulfo-NHS-based approaches and represents a solid basis for further research into the controlled conjugation of tag-free, non-functionalized native-like Env trimers on the surface of liposomes, and other nanoparticles.

Quantities of Intra- and Extracellular DNA Reveal Information About Activity and Physiological State of Methanogenic Archaea.

Although being a common aim of many microbial ecology studies, measuring individual physiological conditions of a microbial group or species within a complex consortium is still a challenge. Here, we propose a novel approach that is based on the quantification of sequentially extracted extracellular (exDNA) and intracellular DNA (iDNA) and reveals information about cell lysis and activity of methanogenic archaea within a biogas-producing microbial community. We monitored the methane production rates of differently treated batch anaerobic cultures and compared the concentrations of the alpha subunit of the methyl coenzyme M reductase gene of methanogenic archaea in extracellular and intracellular DNA fractions and in the classically extracted total DNA pool. Our results showed that this fine-tuned DNA approach coupled with the interpretation of the ratio between free exDNA and iDNA considerably improved microbial activity tracking compared to the classical extraction/quantification of total DNA. Additionally, it allowed to identify and quantify methanogenic populations that are inactive and those that are strongly influenced by cell lysis. We argue that despite the need of further studies, this method represents a novel approach to gain specific physiological information from a complex environmental sample and holds the potential to be applied to other microbes of interest.

Deep optical character recognition: a case of Pashto language

Over the past decades, text recognition technologies have focused immensely on noncursive isolated scripts. A text recognition system for the cursive Pashto script will serve as a great contribution, allowing the traditional, cultural, and educational Pashto literature to be converted into machine-readable form. We propose the use of deep learning architectures based on the transfer learning for the recognition of Pashto ligatures. For recognition analysis and evaluation, the ligature images in the dataset are preprocessed by data augmentation techniques, i.e., negatives, contours, and rotated to increase the variation of each sample and size of the original dataset. Rich feature representations are automatically extracted from the Pashto ligature images using deep convolution layers of the convolution neural network (CNN) architectures using fine-tuned approach. Pretrained CNN architectures: AlexNet, GoogleNet, and VGG (VGG-16 and VGG-19) are used for classification by feeding the extracted features to a fully connected layer and a softmax layer. The proposed deep transfer-based learning has achieved phenomenal recognition rates for Pashto ligatures on benchmark FAST-NU Pashto dataset. An accuracy of 97.24%, 97.46%, and 99.03% is achieved using AlexNext, GoogleNet, and VGGNet architectures, respectively.

Deep Convolution Neural Network for Big Data Medical Image Classification

Deep learning is one of the most unexpected machine learning techniques which is being used in many applications like image classification, image analysis, clinical archives and object recognition. With an extensive utilization of digital images as information in the hospitals, the archives of medical images are growing exponentially. Digital images play a vigorous role in predicting the patient disease intensity and there are vast applications of medical images in diagnosis and investigation. Due to recent developments in imaging technology, classifying medical images in an automatic way is an open research problem for researchers of computer vision. For classifying the medical images according to their relevant classes a most suitable classifier is most important. Image classification is beneficial to predict the appropriate class or category of unknown images. The less discriminating ability and domain-specific categorization are the main drawbacks of low-level features. A semantic gap that exists between features of low-level as machine understanding and features of human understanding as high-level perception. In this research, a novel image representation method is proposed where the algorithm is trained for classifying medical images by deep learning technique. A pre-trained deep convolution neural network method with the fine-tuned approach is applied to the last three layers of deep neural network. The results of the experiment exhibit that our method is best suited to classify various medical images for various body organs. In this manner, data can sum up to other medical classification applications which supports radiologist's efforts for improving diagnosis.

A Deep Learning-Based Hybrid Framework for Object Detection and Recognition in Autonomous Driving

As a key technology of intelligent transportation system, the intelligent vehicle is the carrier of comprehensive integration of many technologies. Although vision-based autonomous driving has shown excellent prospects, there is still a problem of how to analyze the complicated traffic situation by the collected data. Recently, autonomous driving has been formulated as many tasks separately by using different models, such as object detection task and intention recognition task. In this study, a vision-based system was developed to detect and identity various objects and predict the intention of pedestrians in the traffic scene. The main contributions of this research are (1) an optimized model was presented to detect 10 kinds of objects based on the structure of YOLOv4; (2) a fine-tuned Part Affinity Fields approach was proposed to estimate the pose of pedestrians; (3) Explainable Artificial Intelligence (XAI) technology is added to explain and assist the estimation results in the risk assessment phase; (4) an elaborate self-driving dataset that includes several different subsets for each corresponding task was introduced; and (5) an end-to-end system containing multiple models with high accuracy was developed. Experimental results proved that the total parameters of optimized YOLOv4 are reduced by 74%, which satisfies the real-time capability. In addition, the detection precision of the optimized YOLOv4 achieved an improvement of 2.6% compared to the state-of-the-art.

An Effective and Efficient Dynamic eMBMS Multicast Grouping Scheduling Algorithm in MBSFNs for Public Safety Scenarios.

The LTE (Long-Term Evolution) eMBMS (evolved Multimedia Broadcast Multicast Service) technology employed in MBSFNs (Multicast Broadcast Single Frequency Networks) has been shown to be capable of considerably increasing the capacity of serving public safety users under group communication scenarios. However, due to its own limitations, a more fine-tuned scheduling approach is needed in order to fully utilize the strengths of eMBMS multicast. In this work, we first identify and analyze several overlooked challenges for scheduling multicast traffic in MBSFNs. Then we develop an effective and efficient dynamic scheduling algorithm for eMBMS multicast in time and frequency varying channel. The proposed algorithm leverages the advantages of both multicast and unicast schemes via user grouping; and through extensive simulations, is shown to be capable of significantly enhancing the mission critical performances under both best effort and guaranteed bit rate delivery models. We also show the resiliency of our proposed algorithm by applying it onto various network and user deployment scenarios. Our consistent results further prove that LTE eMBMS in MBSFN is a key solution in overcoming limitations in near future public safety networks.

Rethinking the value of unpaid care work: lessons from participatory visual research in central Tanzania

The contemporary international development agenda addresses several important gender issues, including the contribution of unpaid care work (UCW) to human well-being. The inclusion of UCW into the mainstream policy debate is undoubtedly a major milestone in the history of feminist scholarship and activism. However, we argue that the universalistic and capitalocentric assumptions laden in the dominant policy discourse belie the diversity of the lived experiences and subjective meanings of UCW often performed by women and girls in different cultural and geographical contexts, particularly in the predominantly agrarian global South. We draw on participatory and visual ethnographic fieldwork to show that rural women in Tanzania perceive UCW as an experience that entails not only physical toil and drudgery, but also positive emotions of joy, satisfaction, and fulfilment, which are integral to affirming their self-perceived identities and roles as farmers and mothers in their communities. These material, affective, and symbolic dimensions of UCW emerge from agrarian women’s situated knowledges and experiences of ensuring household social reproduction on and with the land, as well as the gender relations and seasonal dynamics that shape the organization of work tasks in agrarian landscapes. To achieve transnational gender justice, we suggest that a more fine-tuned and nuanced approach to understanding the variability and complexity of care work as practiced and perceived by heterogeneous groups of women (and men) in particular places and times is needed.

Using Empathic Design as a Tool for Urban Sustainability in Low-Resource Settings

Architectural design plays a crucial role in sustainable city development. In fast-growing cities in developing countries, it can be a challenge to reach sustainable results. In this paper, we propose the use of Empathic Design, borrowed from the human-centered design field, as one means to support the work of architects and other stakeholders in these settings. To investigate aspects in which this method could be helpful, we have synthesized two existing sustainability models and applied them to three examples of affordable housing from different low-resource settings. After analysis of the examples, we propose a model with an equal balance between the four different dimensions of sustainability—environmental, economic, social, and cultural—where the aspects that need inhabitant engagement are highlighted. We argue that, to be able to hold the balance between the diverse dimensions of sustainability, the architect needs to understand in-depth the living conditions of people for whom he or she is designing. This calls for a fine-tuned participatory approach when designing in low-resource settings. It may not always be easy to reach this level of participation, but we propose that it can be achieved when the architecture is created through empathic involvement. The use of Empathic Design methods throughout the design process thus supports the endeavor towards sustainable results.

Put My Skills to Use? Understanding the Joint Effect of Job Security and Skill Utilization on Job Satisfaction Between Skilled Migrants and Australian Born Workers in Australia

The topic of skilled migrants has gained importance in the past decade as they are increasingly becoming one of the main drivers for labor supply in developed countries like Australia. Although there is research on skilled migrants, most have been studied from the perspectives of (un)employment, wage and over-education. Some evidence suggests that skilled migrants are often less satisfied with their job compared to their local counterparts, yet little is known about why these differences exist. Using a nationally representative sample of Australian workers, we examine how two important job characteristics, job security and skill utilization, exert their differential interaction effect on job satisfaction for skilled migrants and Australian born workers. We found a differential moderation effect between job security and skill utilization for skilled migrants and Australian born workers. For skilled migrants, high job security did not lead to positive reaction (i.e., job satisfaction), as this effect was dependent on their skill utilization; while such moderation effect was not present for Australian born workers. This study highlights the need to take a more fine-tuned approach by understanding target sample groups (e.g., skilled migrants) when study the relationship between key job characteristics and job satisfaction. Furthermore, it highlights the importance for organizations to revisit their human resource management strategies and policies to recognize the needs for enhancing skill utilization for skilled migrants.

A single-variate building energy signature approach for periods with substantial solar gain

The use of regression analysis for the identification of building performance parameters based on measurements is often difficult due to collinearity between the outdoor temperature and the global solar radiation (S). This study proposes a method to overcome this issue. The proposed method is based on using the seasonal symmetry of S to pair data from time-periods equidistant from the winter solstice. In addition, a method to utilize synthetic data to fine-tune the paired-data approach is presented. To evaluate the paired-data approach, two years data from a multifamily building in Umeå was used to estimate the heat loss factor (air-to-air transmission including air leakage). The results were compared with results obtained when S was very low (S≈0). It was found that, the fine-tuned paired-data approach resulted in a modest deviation in the heat loss factor with an average absolute deviation of 4.0%. The small deviation indicates that the paired-data approach can extend the use of single-variate regression models for accurate identification of heat loss factors to situations where the solar gain is substantial. The paired-data approach was also used to calibrate a commercial energy building simulation tool.

Initial assessment of the benefits of implementing pharmacogenetics into the medical management of patients in a long-term care facility.

The health care costs associated with prescription drugs are enormous, particularly in patients with polypharmacy (taking more than five prescription medications), and they continue to grow annually. The evolution of pharmacogenetics has provided clinicians with a valuable tool that allows for a smarter, more fine-tuned approach to treating patients for a number of clinical conditions. Applying a pharmacogenetics approach to the medical management of patients can provide a significant improvement to their care, result in cost savings by reducing the use of ineffective drugs, and decrease overall health care utilization. AltheaDx has begun a study to look at the benefits associated with incorporating pharmacogenetics into the medical management of patients who are on five or more medications. Applying pharmacogenetic guided PharmD recommendations across this patient population resulted in the elimination and/or replacement of one to three drugs, for 50% of the polypharmacy patient population tested, and an estimated US$621 in annual savings per patient. The initial assessment of this study shows that there is a clear opportunity for concrete health care savings solely from prescription drug management when incorporating pharmacogenetic testing.

Performance implications of multinationality: moderating role of foreign market entry mode

Purpose – Previous theoretical explanations and empirical analyses of the multinationality-performance relationship have produced mixed arguments and results. Linear and curvilinear relationships have been theorized and confirmed empirically. This study aims to reconcile earlier mixed findings by incorporating the role of entry mode choice as a moderating variable. Design/methodology/approach – Extensive literature survey is conducted in order to provide an explanation for the different performance patterns of internationalization via licensing, joint venture and wholly owned subsidiary. Findings – Based on the assumptions of Transaction Cost Economics and Organizational Capability perspectives, several research propositions about the multinationality-performance relationship are formulated, which provide a more fine-tuned approach to the performance implications of multinationality. Originality/value – This paper contributes to multinationality-performance debate by showing that the impact of multination...

It's Not Just Structural: Social Movements are not Homogenous Responses to Structural Features, but Networks shaped by Organisational Strategies and Status

Political opportunity structures are often used to explain differences in the characteristics of movements in different countries on the basis of the national polity in which they exist. However, the approach has a number of weaknesses that are outlined in this article. The article especially stresses the fact that such broad-brush approaches to political opportunity structures fail to account for the different characteristics of movement organisations within the same polity. The article therefore recommends using a more fine-tuned approach to political opportunities, taking into account that the strategies and status of organisations affect the real political opportunities they face. This fine-tuned approach is used to predict how the status and strategy of environmental organisations might influence the extent to which different types of environmental organisations in the UK network with one-another. We find that organisations that face an open polity - those with a moderate action repertoire and a constructive relationship with government institutions - tend not to cooperate with those with a radical action repertoire and negative relations with government institutions. On the other hand, those that vary their action repertoires, and which have variable status according to the issues involved or campaign targets, have a much broader range of network links with other types of organisations. Thus, there is much more diversity in types of environmental organisation in the UK than the broad-brush to political opportunity structures would account for. Nonetheless, it does seem that environmental organisations are aware of how their own behaviours might influence (non-structural) political opportunities, and that they mould their strategies and networking patterns around this awareness.

Docenten onder druk? De invloed van schoolgrootte op de arbeidssatisfactie van docenten in het voortgezet onderwijs

Teachers under Pressure? The influence of school size upon teacher job satisfaction in Dutch secondary education Teachers under Pressure? The influence of school size upon teacher job satisfaction in Dutch secondary education In the last decades, a visible and significant upscaling of educational organisations can be witnessed in nearly all Dutch educational sectors. The rise of large-scaled public service organisations is related with increased interest in the quality and 'humanity' of education. Many discussions about large scale education are connected with supposed problems concerning teacher work within schools. It is regularly taken for granted that scale expansion puts teaching staff under pressure, and it has become usual to blame school size for several problems (demotivated professionals, failing quality of education). However, these assumptions are highly controversial. The research of presented in this article intends to answer the question what influence school size has upon teacher job satisfaction in Dutch secondary education. Based on a secondary data-analysis and case study research, we demonstrate that the scale of educational organisations does not necessarily say a lot. In contrast with prevailing views about school size, working in large secondary schools does not by definition harm teacher job satisfaction. On the basis of these findings, we propose a more fine-tuned approach, which attends the organisational context of teachers' work and other factors which affect teacher job satisfaction.

Deconstructing the Library: Reconceptualizing Collections, Spaces and Services

ABSTRACT In the digital environment, we still have resources, staff, and facilities that combine in various ways to acquire and provide information. These recombinations challenge traditional definitions of library organization. Students and faculty now have many options for conducting their work, of which the traditional library is only one; the future of libraries—and librarians—will thus be in our ability to differentiate ourselves through unique and value added features. The library is not a single static entity, it is becoming a suite of services through which users locate, use, and (re)create research materials. By analyzing the intersection of factors such as subject, level of user, type of need, and proximity, we can develop a fine-tuned approach and customize services across the spectrum of physical, electronic, human, and material resources.