Enormous Labor Research Articles

Hyperparameter Optimization for Transfer Learning-based Disease Detection in Cassava Plants

Cassava is quite possibly the most widely recognized staple food crop. It is a nutty-flavored, starchy root vegetable that is a primary energy source and carbs for individuals. During crop cultivation, cassava plant infections can influence the leaf and root, bringing about a tremendous loss to the harvest and financial market esteem. Hence, it is vital to detect diseases in cassava plants. But it requires enormous labor, longer time planning, and thorough plant-specific knowledge. If disease detection is possible at the initial stages, then actions can be taken on time. Hence, there is a need to develop automatic detection methods for monitoring different parts of cassava plants. This study evaluates the efficiency of applying transfer learning to the pre-trained models for identifying diseases in cassava plants. The pre-trained EfficientNet model detects the disorders using data augmentation, fine-tuning the hyperparameters, cross-validation, and transfer learning. The experimentation is done with the cassava dataset provided by Kaggle, which contains cassava plant leaf images belonging to five classes. An experimental investigation shows that EfficientNet with transfer learning attains up to 89% accuracy. The effect of transfer learning is significant; consider getting the results of high accuracy and less dispersion; in very few cases, the model forecasts the wrong class labels. The outcomes give a promising strength to the objective of this work, i.e., a model trained explicitly for agriculture with transfer learning can assist the farmers with highly accurate results during farming to get a high yield.

Self-Supervised Correlation Learning for Cross-Modal Retrieval

Cross-modal retrieval aims to retrieve relevant data from another modality when given a query of one modality. Although most existing methods that rely on the label information of multimedia data have achieved promising results, the performance benefiting from labeled data comes at a high cost since labeling data often requires enormous labor resources, especially on large-scale multimedia datasets. Therefore, unsupervised cross-modal learning is of crucial importance in real-world applications. In this paper, we propose a novel unsupervised cross-modal retrieval method, named Self-supervised Correlation Learning (SCL), which takes full advantage of large amounts of unlabeled data to learn discriminative and modality-invariant representations. Since unsupervised learning lacks the supervision of category labels, we incorporate the knowledge from the input as a supervisory signal by maximizing the mutual information between the input and the output of different modality-specific projectors. Besides, for the purpose of learning discriminative representations, we exploit unsupervised contrastive learning to model the relationship among intra- and inter-modality instances, which makes similar samples closer and pushes dissimilar samples apart. Moreover, to further eliminate the modality gap, we use a weight-sharing scheme and minimize the modality-invariant loss in the joint representation space. Beyond that, we also extend the proposed method to the semi-supervised setting. Extensive experiments conducted on three widely-used benchmark datasets demonstrate that our method achieves competitive results compared with current state-of-the-art cross-modal retrieval approaches.

Active learning for efficient analysis of high-throughput nanopore data.

As the third-generation sequencing technology, nanopore sequencing has been used for high-throughput sequencing of DNA, RNA, and even proteins. Recently, many studies have begun to use machine learning technology to analyze the enormous data generated by nanopores. Unfortunately, the success of this technology is due to the extensive labeled data, which often suffer from enormous labor costs. Therefore, there is an urgent need for a novel technology that can not only rapidly analyze nanopore data with high-throughput, but also significantly reduce the cost of labeling. To achieve the above goals, we introduce active learning to alleviate the enormous labor costs by selecting the samples that need to be labeled. This work applies several advanced active learning technologies to the nanopore data, including the RNA classification dataset (RNA-CD) and the Oxford Nanopore Technologies barcode dataset (ONT-BD). Due to the complexity of the nanopore data (with noise sequence), the bias constraint is introduced to improve the sample selection strategy in active learning. Results: The experimental results show that for the same performance metric, 50% labeling amount can achieve the best baseline performance for ONT-BD, while only 15% labeling amount can achieve the best baseline performance for RNA-CD. Crucially, the experiments show that active learning technology can assist experts in labeling samples, and significantly reduce the labeling cost. Active learning can greatly reduce the dilemma of difficult labeling of high-capacity nanopore data. We hope active learning can be applied to other problems in nanopore sequence analysis. The main program is available at https://github.com/guanxiaoyu11/AL-for-nanopore. Supplementary data are available at Bioinformatics online.

Federated learning based multi‐task feature fusion framework for code expressive semantic extraction

AbstractUsing multi‐task learning to extract code features can effectively increase the information of the features. However, the existing multi‐task learning methods mainly have two limitations: (1) They cannot extract enough code‐related information or only extract similar semantic features. Similar multi‐task makes the information in the features increased insufficiently. However, the high difference multi‐task is challenging to converge. (2) They cannot train multi‐task on heterogeneous datasets. In standard multi‐task training, we need to label all tasks for all data, which consumes enormous labor. To solve the above limitations, we select two high difference tasks, the cross‐language code completion task and variable misuse task, to extract expressive semantic code features. We propose an attention‐based feature fusion module to merge information among high difference tasks, avoiding the convergence dilemma of standard multi‐task learning. We propose a federated learning framework, extracting semantic information and using the feature fusion module to integrate multi‐task information among single labeled datasets. We experiment on C# and Python datasets for the code completion and variable misuse tasks. The results show that the performance of fusion features by FedMTFF improved by up to 22.6% and 15.1% compared to single tasks. We use FedMTFF to perform four cross‐language multi‐task features fusion, exceeding the current best baseline by 24.1%.

The Importance of Utilizing Gravity Tensor Invariant in Determining Hydrocarbon Locations

The region, which covers the northeast of Syria and the northwest of Iraq in the south of the Bitlis-Zagros Suture Belt, which was formed as a result of the collision of the Arabian and Eurasian plates during the Middle Miocene and Early Late Miocene, has a very important position in the world in terms of hydrocarbons. Hydrocarbon exploration activities require enormous labour and cost. It is aimed to show that the locations of the known hydrocarbon fields in the region coincide with the maximum values of the gravity tensor invariant with this study. In this way, it is aimed to show that both the working area can be narrowed and the cost can be reduced by applying this method beforehand in similar regions. For this purpose, World Gravity Map 2012 global model gravity data, which is one of the most up-to-date satellite-based gravity data used by many researchers recently, was used. First of all, the spherical free air and complete spherical Bouguer gravity data were analyzed and then the potential was obtained by taking the vertical integration of the complete spherical Bouguer data. Tensors were calculated by using this potential data and gravity tensor invariant was calculated from tensors. When the gravity tensor invariant map is examined, it is observed that the areas with the maximum positive values observed and the locations of the known oil wells are mostly compatible throughout the area. As a result, it would be very beneficial to apply this method for the region first for such high-cost studies.

A semi-supervised learning method for hyperspectral imagery based on self-training and local-based affinity propagation

ABSTRACT In hyperspectral remote sensing, the classification of hyperspectral imagery is an important issue of concern. However, obtaining sufficient labelled samples for the classification is hard work and consumes enormous labour and material resources. To solve this problem, many scholars make their efforts to the semi-supervised learning method. The self-training model is a popular semi-supervised learning method, which trains a classifier using a few labelled samples and many unlabelled samples in an iterative manner. Nevertheless, the self-training model may provide misclassification as adding incorrect unlabelled samples. This paper proposes a novel semi-supervised learning method, which attempts to introduce the affinity propagation and spatial constraint to self-training model. In the proposed algorithm, a self-training model is employed to expand the labelled sample set. The affinity propagation is applied to select a group of reliable unlabelled samples. To further protect the reliability of unlabelled samples, the selection process of self-training model is constrained in the spatial local area of the labelled samples. In the iteration, the reliable unlabelled samples are selected and added to the labelled sample set. After that, the classifier is trained by the updated labelled samples. The algorithm finishes as the iterations exceed the predetermined value or the labelled sample set is unchanged. Different from the traditional self-training-based semi-supervised learning method, there are mainly three improvements. First, the selection of reliable unlabelled samples is conducted by the affinity propagation, which can effectively improve the reliability of the unlabelled samples. Second, the spatial constraint is introduced to restrict the selection process of self-training model, which can protect the reliability of unlabelled samples and alleviate the computational complexity. Finally, the spectral correlation angle is applied to construct the similarity matrix of affinity propagation, so that the algorithm is more suitable for hyperspectral imagery. A comparative analysis was conducted with classical self-training semi-supervised algorithms to test the effectiveness of the method. It was found that the proposed method provided higher overall accuracy, average accuracy and kappa coefficient (κ) than the competitors.

Weeds Impose Unique Outcomes for Pests, Natural Enemies, and Yield in Two Vegetable Crops.

Weed management requires enormous labor investments from vegetable farmers, yet crops vary in how much weed pressure they can tolerate without yield loss. Moreover, until weeds reach a point where they threaten yield or approach seed production, they can increase biodiversity and provision food and habitat to attract predatory insects. In two related field experiments, we quantified impacts of weed presence and diversity on pests, predators, and biocontrol of both weed seeds and insect prey. We also measured yields of two vegetables that vary in competitiveness (eggplants and turnips) across two weed management treatments (weedy and weed-free), to determine productivity costs of tolerating weeds. Allowing weeds to grow adjacent to rows of eggplants increased abundances of predators and reduced pests. Surprisingly, relaxing weed management came at no cost to eggplant yield. In contrast, tolerating weeds in turnips had strong yield costs, and did not benefit predators or decrease pest pressure. On both crops, pests declined as weed diversity increased. Yet, weed treatments had no impact on consumption of weed seeds or sentinel prey by soil-surface insects, which were dominated by red imported fire ants. Our results suggest that highly competitive crops might benefit from stronger natural pest control when weeds are less-aggressively managed. However, herbivores and predators had unique responses to weeds that were crop-specific. To help farmers allocate limited weed management labor resources, future work should examine the relative competitiveness of a wider variety of vegetables over a gradient of weed pressure while measuring corresponding impacts on pest control.

Performance of deep learning vs machine learning in plant leaf disease detection

Plants are recognized as essential as they are the primary source of humanity's energy production since they are having nutritious, medicinal, etc. values. At any time between crop farming, plant diseases can affect the leaf, resulting in enormous crop production damages and economic market value. Therefore, in the farming industry, identification of leaf disease plays a crucial role. It needs, however, enormous labor, greater preparation time, and comprehensive plant pathogen knowledge. For the identification of plant disease detection various machine learning (ML) as well as deep learning (DL) methods are developed & examined by various researchers, and many of the times they also got significant results in both cases. Motivated by those existing works, here in this article we are comparing the performance of ML (Support Vector Machine (SVM), Random Forest (RF), Stochastic Gradient Descent (SGD)) & DL (Inception-v3, VGG-16, VGG-19) in terms of citrus plant disease detection. The disease classification accuracy (CA) we received by experimentation is quite impressive as DL methods perform better than that of ML methods in case of disease detection as follows: RF-76.8% > SGD-86.5% > SVM-87% > VGG-19–87.4% > Inception-v3–89% > VGG-16–89.5%. From the result, we can tell that RF is giving the least CA whereas VGG-16 is giving the best in terms of CA.

A generalized predictive model for TiO2–Catalyzed photo-degradation rate constants of water contaminants through artificial neural network

Titanium dioxide (TiO2) is a well-known photocatalyst in the applications of water contaminant treatment. Traditionally, the kinetics of photo-degradation rates are obtained from experiments, which consumes enormous labor and experimental investments. Here, a generalized predictive model was developed for prediction of the photo-degradation rate constants of organic contaminants in the presence of TiO2 nanoparticles and ultraviolet irradiation in aqueous solution. This model combines an artificial neural network (ANN) with a variety of factors that affect the photo-degradation performance, i.e., ultraviolet intensity, TiO2 dosage, organic contaminant type and initial concentration in water, and initial pH of the solution. The molecular fingerprints (MF) were used to interpret the organic contaminants as binary vectors, a format that is machine-readable in computational linguistics. A dataset of 446 data points for training and testing was collected from the literature. This predictive model shows a good accuracy with a root mean square error (RMSE) of 0.173.

Books Are Made Out of Books: A Guide to Cormac McCarthy's Literary Influences

Books Are Made Out of Books: A Guide to Cormac McCarthy's Literary Influences

Growth and physiology of basmati rice under conventional and water-saving production systems

ABSTRACTConventionally flooded rice (CFR) requires enormous water and labor inputs. Water scarcity aspires for cultivation of water-saving rice. Growth response and physiology of basmati rice genotypes under the water-saving production systems has not been reported yet. Studies were conducted for 2 years to compare the growth and physiology of three rice cultivars (Super Basmati, Basmati-2000 and Shaheen Basmati), under high (CFR), medium (alternate wetting and drying [AWD]) and low water input (aerobic rice [AR]) systems. Leaf area index, crop growth rate, leaf area duration and dry matter accumulation were higher for AR followed by AWD and CFR, respectively. Shaheen Basmati had a lower growth and relative water contents than Super Basmati and Basmati-2000, probably due to its shorter stature and shorter life cycle. Photosynthetic rate and stomatal conductance of rice cultivars in the different production were affected only at reproductive stage. Basmati-2000 grown as AR had the highest photosynthetic rate followed by the same cultivar under AWD. The results of this study provide us an idea that basmati cultivars can attain a high growth and development with low water input. This would be helpful to grow rice successfully under water-short rice-growing environments.

A City for Children: Women, Architecture, and the Charitable Landscapes of Oakland, 1850-1950 by Marta Gutman

Book Reviews 197 children in the design of space and places for them both in the community and in structured spaces, where children spend much of their time. Policy makers and educators could learn much from reading these chapters. Indeed, if we continue to direct and shape the spaces our children can inhabit, then understanding the dissonances between the purpose and use of spaces for children, and the value of embodied, affective and just children’s spaces is essential. References Bourdieu, P. (1977). Outline of a theory of practice. (Richard Nice, Trans.). Cambridge: Cambridge University Press. MacRae, C. (2011). Making Payton’s rocket: Heterotopia and lines of flight. International Journal of Art & Design Education, 30(1), 102–112. Ingold, T. (2007). Lines: A brief history. London: Routledge. Tilley, C. (2004). The materiality of stone: Explorations in landscape phenomenology. Oxford: Berg. Review by Kumara Ward Dr. Kumara Ward is an academic at Western Sydney University teaching curriculum and Pedagogy in the Early Childhood Education program and a member of the Centre for Educational Research – Sustainability Strand. Her teaching, doctorial studies and current research focus on the symbiosis between the arts and the natural and built worlds and ways in which they promote embodied interaction and multifaceted ways of knowing our complex human and other than human common worlds. A City for Children: Women, Architecture, and the Charitable Landscapes of Oakland, 1850-1950 Marta Gutman (2014) Chicago and London: University of Chicago Press, 454 pages. $45.00; ISBN: 978-0-226-311289 Idealism followed industry into Oakland during 1870. The arrival in California of the transcontinental railroad in November 1869 helped hasten the growth of a settlement some 10,500 strong into one playing host, just 20 years later, to nearly 50,000 residents and all manner of attendant social concerns. Female charitable enterprise fanned out across the expanding city, with the Ladies’ Relief Society and sister organizations building from scratch upon foundations of religious Book Reviews 198 determination, expertise from neighboring San Francisco and no little local resolve. Into this “charitable landscape” steps Marta Gutman, scholar, architect and our expert chaperone to a world now mostly lost. A City for Children covers a hundred years and more of Oakland’s history, taking the reader up to the destruction—facilitated by an ex-army Sherman tank, no less— of the New Century Recreation Center, a neighboring playground, and public school. That recreation center and playground, together with kindergartens, orphanages, homes for the destitute, and daycare facilities, formed nodal points in a web of civic philanthropy. It is the very great achievement of this book to recover, reimagine and scrutinize this landscape made for the benefit of children by redoubtable women including dressmaker Rebecca McWade and teacher Elizabeth Betts. Eschewing the tendency of historians of childhood to focus on places for children in isolation, Gutman knits together the entire charitable city as it would have appeared to its sponsors and beneficiaries. To achieve this, the author draws on archival evidence ranging from oral histories to government reports and photographs, all interpreted with acuity. Particularly noteworthy are Gutman’s remarkable (and self-fashioned) neighborhood maps depicting the scale, layout and crucial proximity of institutions charitable and religious. Based on details gleaned from insurance records and (as the author states very modestly) “information culled from city directories, tax records, and fieldwork” (see, for instance, page 107), these maps represent an enormous labor of love and help excavate a layer of city life now obscured. Such painstaking attention to place is characteristic of a study many years in the making. Standing out as memorable landmarks in Gutman’s landscape are buildings intended for one activity and appropriated—or “repurposed”—for another. The chapter on “The Saloon That Became a School” exemplifies the author’s arguments regarding activists’ drive for identity to flow from spatial configuration. For example, Elizabeth Betts’ West Oakland Free Kindergarten occupied a site facing the rail terminus and intended to promote sobriety and particular forms of socialization among the many immigrant children arriving into the city. The “metaphor and practice of salvage” (p. 30) also characterized women, either privileged or more modest in means, who transformed private homes into charitable spaces, a process...

대규모 IT 프로젝트를 위한 실용적 행태관리 방안

본 기술의 진화는 역사상 유래 없는 속도로 사회와 기업의 변화를 가속하고 있고 요구되는 대응을 위해 대형 IT 프로젝트가 긴박하게 추진되고 있다. 고도의 기술과 대규모 업무를 다루고, 막대한 인력과 비용이 투입되는 대규모 프로젝트는 인간 심리와 복잡계 이론에 기반한 전략적 갈등관리가 요구되나, 기존의 관리방법은 이에 부응하는 실용적 기제를 제시하지 못하는 실정으로 프로젝트 실패가 증가하고 있다. 본 연구는 적절한 해법을 찾기 위해 프로젝트 실패 현황과 성과 요인, 프로젝트 리더십에 대해 관련 연구하였고, 대형 IT 프로젝트의 복잡성과 갈등구조에 주목하여 기존의 프로젝트 관리방안에서는 간과하였던 심리학적, 행동경제학적 접근을 접목하였다. 논문에서 제시한 프로젝트 행태관리 방안은 대규모 IT 프로젝트의 주요 실패 행태들을 근원적이고 선제적으로 해소할 수 있는 프로젝트 단계별 행태관리 전략과 실천적 프랙티스를 제시한다. 해당 현장 전문가 및 실무자 그룹의 설문조사를 통해 제안된 행태관리 방안의 유효성에 대한 기대를 검증하였고, 본 방안을 대형 프로젝트 관리에 도입함으로써 보다 안정적인 프로젝트 성공 기반을 확보할 수 있다. Technology developments accelerate the change of the society and companies with unparalleled pace, so large scaled projects are in progress urgently. Large scaled projects deal with high technologies and massive tasks, and these need enormous labor force and costs, it is necessary to manage strategic conflicts based on human psychology and complex system theory. However, traditional management plan fails to suggest practical mechanism for successful projects. We study failure cases and key elements, leadership for projects, and suggest project management plan with psychological and behavioral economy approach by focusing on complexity and conflict structure on large scaled projects. Project behavior management plan, suggested in this study, is designed with phased strategy and practice to solve the failure of large scaled IT projects originally and preemptively. We verified the effectiveness of behavior management plan suggested by investigating experts and working groups, and stable projects are to be progressed with introducing this plan on large-scaled projects.

China's town and village enterprises and its implications for Sub-Saharan Africa

The successful development of China's town and village enterprises (TVEs) has significantly absorbed enormous rural labour surplus during the agricultural transformation and the process of urbanisation between 1984 and 2014. TVEs played an important role in poverty reduction and increasing income for rural labours in China, when markets were dysfunctional and institutions were changing during the 1980s and 1990s. Assessments have concluded that Sub-Saharan Africa (SSA) has similar challenges and conditions with respect to agricultural transformation and managing urbanisation to those of China in the 1980s-1990s. This research uses interviews and case studies gathered mainly in China in 2014. The research finds that dramatic development of TVEs helped China to get rid of the development dilemma of a dualistic economic structure in rural and urban areas, which can bring valuable lessons for SSA. The definition of TVEs in this article has been extended to inclusive local enterprise organisations. This research suggests feasible policies to promote local social enterprises with inclusive development entrepreneurship in SSA.

Supplier networks in the German aircraft industry during World War II and their long-term effects on West Germany's automobile industry during the ‘Wirtschaftswunder’

Reconstructing the complex supplier network of the famous JU 88 air armament programme, this article shows that outsourcing activities increased considerably in wartime Germany. The resulting inter-firm division of labour did not lead only to a quite effective protection of the German aircraft production against Allied air raids but also contributed to enormous labour productivity growth in most stages of the production process. Even though aircraft production was prohibited in post-war Germany, this supplier network survived and became the backbone of the most spectacular symbol of West Germany's economic rebirth: the automobile industry.

Working poverty in Michigan, 1998/1999 and 2007/2008: Changes in the magnitudes and policy and socio-demographic determinants

Michigan has undergone enormous labor market changes since the 1990s affecting employment, income, and poverty. This paper examines changes in poverty among working families and their policy-related and socio-demographic determinants between 1998/1999 and 2007/2008 in Michigan. Findings suggest the rates of ‘poverty’ and ‘near poverty’ to be between 5 and 19% among working families, with slightly higher rates for the latter period. Public transfers combining taxes and means-tested supports, albeit making some impact among poor families with children, were unable to lower these rates. While the major socio-demographic characteristics of poverty and near poverty including large families with children and young, never married, single mother, and immigrant householders apply to working families, the roles of gender, race, marital status, and education manifest through many policy-related variables such as work hours, wages, and transfer incomes. These findings have important implications for understanding working poverty in Michigan and beyond.

Production beyond the Horizon of Consumption: Spatial Fixes and Anti-sweatshop Struggles in the Global Athletic Footwear Industry

Transnational outsourcing makes it possible for Western companies to access the enormous labour reserves in countries such as China, India or Bangladesh without entering into formal (contractual) relations with these workers. It provides global buyers with an opportunity to disassociate themselves from (labour-intensive) production activities, and thereby from struggles over wages, environment, conditions of work and reproduction. In this context, workers are increasingly treated as a subcontracted component rather than a fixture as part of employer organisations. Drawing upon concepts derived from human and labour geography, this paper takes one labour-intensive sector—athletic footwear—as a case study of how the transnationalisation of production has resulted in a sweatshop model, while simultaneously transforming the athletic footwear industry into a site of social struggle. This paper looks at how global buyers have used relocation as a spatial fix for crises of profitability and labour control, but also pays attention to countervailing tendencies such as the emergence of a global anti-sweatshop movement, and the influence of codes of conduct on working conditions, which adds an extra-local dimension to labour relations.

RISING WAGES: HAS CHINA LOST ITS GLOBAL LABOR ADVANTAGE?

We document dramatic rising wages in China for the period 1978–2007 based on multiple sources of aggregate statistics. Although real wages increased seven-fold during the period, growth was uneven across ownership types, industries and regions. Over the past decade, the wages of state-owned enterprises have increased rapidly and wage disparities between skill-intensive and labour-intensive industries have widened. Comparisons of international data show that China's manufacturing wage has already converged to that of Asian emerging markets, but China still enjoys enormous labour cost advantages over its neighbouring developed economies. Our analysis suggests that China's wage growth will stabilize to a moderate pace in the near future.

A Study for Producing 3D Model Space from 2D Digital Picture using Photogrammetry Technology

Three-dimensional CAD and CG, which make use of advanced computer technology, has lately attracted considerable attention. Also attracting attention is photogrammetric technology to acquire three-dimensional coordinate data from two-dimensional image data recorded by digital cameras. Photographing and the determination of measuring points are more difficult even for experts in ground photogrammetry, for which the demand is increasing, as compared with the same tasks in aerial photogrammetry which is used for the preparation of maps and so on. Accordingly it takes enormous labor and time to obtain accurate three-dimensional coordinate data by ground photogrammetry and, therefore, it has hardly been put to practical use yet. Under the circumstances, the object of the present study was to develop a system which enables us, without requiring expensive apparatuses or expertise, to create 3D model spaces easily and quickly just by taking two-dimensional digital photographs.

Rising Wages: Has China Lost its Global Labor Advantage?

We document dramatic rising wages in China for the period 1978-2007 based on multiple sources of aggregate statistics. Although real wages increased seven-fold during the period, growth was uneven across ownership types, industries and regions. Since the late 1990s, the wages of state-owned enterprises have increased rapidly and wage disparities between skill-intensive and labor-intensive industries have widened. Comparisons of international data show that China's manufacturing wage has already converged to that of Asian emerging markets, but China still enjoys enormous labor cost advantages over its neighboring developed economies. Our analysis suggests that China's wage growth will stabilize to a moderate pace in the near future.