1,721,568 research outputs found
Parametric and non-parametric approaches for runoff and rainfall regionalization
Full text linkThe information on river flows is important for a number of reasons including; the construction of hydraulic structures for water management, for equitable distribution of water and for a number of environmental issues. The flow measurement devices are generally installed across the workspace at various locations to get data on river flows but due to a number of technical and accessibility issues, it is not always possible to get continuous data. The amount rainfall in a basin area also contributes towards the river flows and intense rainfall can cause flooding. The extended rainfall maps for the study areas to analyze these extreme events can be of great practical and theoretical interest. This thesis can be generally regarded as a work on catchment hydrology and mapping rainfall extremes to estimate certain hydrological variables that are not only useful for future research but also for practical designing and management issues. We analyzed a number of existing techniques available in literature to extend the hydrological information from gauged basin to ungauged basin; and suggested improvements. The three main frontiers of our work are: Monthly runoff regime regionalization, Flow duration curves (FDCs) regionalization and preparing rainfall hazardous maps. The proposed methods of regionalization for runoff regime and FDCs are tested for the basins located in northern Italy; whereas for rainfall extremes, the procedure is applied to the data points located in northern part of Pakista
Utilizzo di Large Language Models in Domini con Pochi Dati: Una Prospettiva Industriale
No full textIn recent years, there has been a significant shift toward online job postings and recruitment portals, allowing candidates to easily upload their data and documents—such as resumes and CVs—for specific job vacancies. These platforms have streamlined the application process for candidates but have also made the screening process more time-consuming and labor-intensive for recruiters. For a single job advertisement, the Human Resources (HR) department may receive a large number of applications.
Natural Language Processing (NLP) tools have the potential to alleviate this burden, saving valuable HR resources by automating parts of the recruitment process. Specifically, automatic information extraction from text data can expedite recruiters’ tasks by rapidly identifying relevant candidate information, such as personal details, work experience, and education. Additionally, soft skills are a critical component that recruiters assess when screening candidates for a particular job profile. A job profile typically includes desired attributes such as educational background, technical and soft skills, and past roles held. By leveraging NLP, recruiters can more effectively match these qualifications to find the ideal candidate for each position.
The primary objective of this thesis is to investigate and improve existing NLP approaches to meet the needs of the industry. We identify several key challenges in this domain, including scarcity of training data, complex information extraction requirements, and a lack of standardized approaches.
To address these challenges, we have enhanced data augmentation techniques for complex information scenarios, particularly in low-resource contexts within Human Resource Management. By leveraging advanced NLP techniques, we generated synthetic data that mirrors the structure and nuances of real-world HR datasets, enriching the data where actual examples are limited. This approach not only increased the diversity and volume of training data but also improved the robustness of downstream models in handling complex and varied inputs. Additionally, we utilized recent Large Language Models (LLMs), such as GPT, to automate data annotation tasks, enabling faster and more accurate labeling of HR-specific information. This integration of LLMs has streamlined the annotation process, providing high-quality labeled datasets with minimal manual effort and making it feasible to train more sophisticated models in low-resource domains
Pitfalls in transboundary Indus Water Treaty: a perspective to prevent unattended threats to the global security
Full text linkWater treaties have played an important role in peaceful resolution of water-related conflicts. Although the mode of negotiation to resolve water-related conflicts may vary from treaty to treaty, a number of structural falls make them unprepared for the future needs. The Indus water treaty is perhaps quoted as the most successful water-sharing mechanism in the recent times. Against all odds, the treaty has fulfilled its job descriptions of being a mechanism providing a moderately reliable framework for the peaceful resolution of water-related conflicts. However, the climate change is quickly eroding that trust. The water-sharing mechanism lacks guidelines to cater the issues related to climate change and basin sustainability which require integrated approach for their addressal. But the structural inflexibility does not encourage the riparian to collaborate and build mutual trust for common good. The riparian countries, within the framework of treaty, attempt to elevate their national interests by deliberately refusing to comply with the treaty clauses in letter and spirit, and even manipulate data to deprive the competing riparian of water. We propose and argue on the need of adopting structurally sound forum for solving water conflicts which will assist in comprehensive policy-making to ensure the sustainability of transboundary water resources. The forum will also provide an opportunity for the riparian to work together towards confidence-building through sharing of real-time hydrological data and further scientific analysis based on that. Conclusively, the shortcomings of the present conflict-resolution method are addressed by encouraging riparian to collaborate at various levels
High altitude vs underwater : An analysis of functional lung capacity / Muhammad Uzair Azim Azman
Full text linkPorter is the one who involved in high altitude condition while scuba divers is the
one who involved in underwater condition. Porters and scuba divers need a very
good lung in order to maintain their performance during their activities and they
usually have greater lung capacity than others because of their adaptation in high
altitude and underwater condition. The purpose of this study is to determine and
compare the functional lung capacity that is Forced Vital Capacity (FVC, Forced
Expiratory Volume in 1 second (FEV1) and FEV1/FVC between Mount Kinabalu
Porters and Sapi Island Scuba Divers. The spirometric parameters were measured at
two different subject which is porters that involve in high altitude and divers that
involved in underwater. The independent variable (IV) is representing the porters and
divers and dependent variable (DV) will be representing the functional lung capacity.
15 porters and 15 divers were tested using spirometer and the FVC, FEV1 and
FEV1/FVC was measured. The data collected have been analyzed using Statistical
Package for Social Science (SPSS) version 19.0.The results show that FVC value
was 0.03 which is less than 0.05. FEV1 value is 0.02 which was also less than 0.05.
While FEV1//FVC (L) value was 0.5 which is more than 0.05. There is a significant
difference effect on FVC and FEV1 between porters and divers. Thus the null
hypothesis for FVC and FEV1 is accepted while the FEV1/FVC (L) showed that
there is no significant difference. This study showed that porters have greater lung
capacity than divers based on their FVC and FEV1 but not in FEV1/FVC (L)
Prompt-Based Data Augmentation Using Contrastive Learning Under Scarcity of Annotated Data
Full text linkGraph Neural Networks for Candidate-Job Matching: An Inductive Learning Approach
No full textThis work introduces a novel graph-based approach to candidate-job matching using Graph Neural Networks (GNNs). We analyzed data from 62 real-world selection processes encompassing 8360 unique candidates and 9532 applications, characterized by extreme class imbalance (95% rejection rate). Our methodology constructs purpose-built bipartite graphs for each candidate-job pair, with 14 nodes representing candidates, jobs, and their respective attributes extracted using Large Language Models. Each graph contains a minimum of 15 edges representing semantic relationships between entities, with edge weights derived from embedding similarity measures. We empirically evaluated five GNN architectures (GCN, MIGNN, GIN, GAT, GraphConv) against standard neural networks across binary and ordinal classification tasks. In binary classification, graph-based approaches consistently outperformed non-graph baselines, with GCN achieving 65.4% balanced accuracy compared to 55.0% for the MLP baseline. GNN models also demonstrated superior minority class detection, with GCN correctly identifying 48.9% of qualified candidates versus only 8.5% for MLP. Statistical analysis revealed that higher recruitment stages correlate with increased graph connectivity, validating our graph construction methodology. While all models struggled with ordinal classification, the explicit modeling of semantic relationships through graph structures enabled effective binary discrimination for candidate screening, offering a promising direction for augmenting human decision-making in recruitment processes while maintaining interpretability
Banco company / Muhammad Uzair Mansor ... [et al.]
Full text linkBeside that we make oat wheat biscuit is very suitable for every level age. The variety of flavor in our biscuit make this biscuit interesting and fun for any occasion. The size of this biscuit is most important because it small so that make the customer easy to bring it together everywhere and every time especially who is busy to eat so to get energy temporary it can eat our biscuit. Our biscuit is make from oats in an oatmeal cookie contain soluble fiber, which reduces your bad cholesterol level and decreases your risk of heart disease. Soluble fiber can also help regulate your blood sugar levels. Our biscuit is concept about healthy sneak so it contain ingredient based on wheat such vitamin, minerals and fiber. Folate is a B vitamin that helps you make energy and it also lowers the risk of certain birth defects. The same oatmeal cookie supplies tiny doses of vitamin A for healthy eyes and vitamin K, which clots your blood. Another good thing is the most notable mineral you'll get from an oatmeal cookie is iron. Iron plays a role in the formation of red blood cells and can also protect your immune system. The same oatmeal cookie delivers small amounts of potassium, which you need for healthy muscles, as well as zinc, which aids in wound healing. Moreover, the simple steps and ingredients give a lot of benefits to our company and we are highly confident that our market can be easily developed and spread in Malaysia. We have make decision to sell the biscuit not only sell but can be as gift by packaging in the jars and the packet look like candy sweet so customer can make a demand to get the biscuit in big amount in achievable price
Predicting Peak Flows in Real Time through Event Based Hydrologic Modeling for a Trans-Boundary River Catchment
No full textInvestigating the hydrological response of an area to adverse climate changes and extreme rainfall events is crucial for managing land and water resources and mitigating the natural hazards like floods. Limited availability of the in situ data, especially in case of Transboundary Rivers, further highlights the need to develop and evaluate decision support systems which may predict the flows in real time using open source rainfall data. This paper presents the study conducted in Chenab River catchment, Pakistan, to develop and evaluate a hydrologic model using HEC-HMS for predicting flows based on TRMM rainfall data. The catchment was analyzed for hydro-morphological properties using SRTM DEM in HEC-GeoHMS. To rely on open source data as much as possible, digital soil map of the world developed by FAO and global land cover map developed by European Space Agency were utilized to compute Curve Number grid data for the catchment. These preliminary data analyses were employed to set initial values of different parameters to be used for model calibration. The model was calibrated for five rainfall events occurred in the rainy seasons of 2006, 2010 and 2013. The calibrated model was then validated for four other rainfall events of similar type in the same years. Consistency in simulated and observed flows was found with percent difference in volume ranging from −6.17 % to 5.47 % and percent difference in peak flows to be in the range of 6.96 % to 7.28 %. Values of Nash-Sutcliffe Efficiency were ranging from 0.299 to 0.909 with an average value of 0.586 for all flow events. The model was found well capable of capturing the hydrologic response of the catchment due to rainfall events and can be helpful in providing alerts of peak flows in real time based on real time/forecasted rainfall data
- …
