A5043237258- Atmospheric aerosols and clouds
- Atmospheric chemistry and aerosols
- Information Retrieval and Search Behavior
- Solar Radiation and Photovoltaics
- Atmospheric Ozone and Climate
- Air Quality Monitoring and Forecasting
- Atmospheric and Environmental Gas Dynamics
- Meteorological Phenomena and Simulations
- Land Rights and Reforms
- Climate variability and models
- Advanced Algebra and Geometry
- scientometrics and bibliometrics research
- Neural Networks and Applications
- Advanced Mathematical Modeling in Engineering
- Advanced Image and Video Retrieval Techniques
- Web Data Mining and Analysis
- Mathematical and Theoretical Analysis
- Art, Politics, and Modernism
- Rangeland Management and Livestock Ecology
- Semantic Web and Ontologies
- Topic Modeling
- Algebraic and Geometric Analysis
- Statistical and numerical algorithms
- Agriculture, Land Use, Rural Development
- Advanced Text Analysis Techniques
University of East Anglia
2023-2024
Accuray (United States)
2024
University of Connecticut
2024
University of St Andrews
1995-2020
University of Reading
2017-2019
Aristotle University of Thessaloniki
2016-2017
Ithaca College
2017
National Observatory of Athens
2009-2016
University of Winchester
2016
Inter-American Development Bank
2015
Search engine click logs provide an invaluable source of relevance information, but this information is biased. A key bias presentation order: the probability influenced by a document's position in results page. This paper focuses on explaining that bias, modelling how depends position. We propose four simple hypotheses about might arise. carry out large data-gathering effort, where we perturb ranking major search engine, to see clicks are affected. then explore which best explains...
This paper describes a simple way of adapting the BM25 ranking formula to deal with structured documents. In past it has been common compute scores for individual fields (e.g. title and body) independently then combine these (typically linearly) arrive at final score document. We highlight how this approach can lead poor performance by breaking carefully constructed non-linear saturation term frequency in function. propose much more intuitive alternative which weights frequencies before...
Abstract A climate data record of global sea surface temperature (SST) spanning 1981–2016 has been developed from 4 × 10 12 satellite measurements thermal infra-red radiance. The spatial area represented by pixel SST estimates is between 1 km 2 and 45 . mean density good-quality observations 13 −2 yr −1 uncertainty evaluated per datum, the median for SSTs being 0.18 K. Multi-annual observational stability relative to drifting buoy within 0.003 K zero with high confidence, despite maximal...
We address the problem of learning large complex ranking functions. Most IR applications use evaluation metrics that depend only upon ranks documents. However, most functions generate document scores, which are sorted to produce a ranking. Hence innately non-smooth with respect due sort. Unfortunately, many machine algorithms require gradient training objective in order perform optimization model parameters,and because non-smooth,we need find smooth proxy can be used for training. present...
Given a terabyte click log, can we build an efficient and effective model? It is commonly believed that web search logs are gold mine for business, because they reflect users' preference over documents presented by the engine. Click models provide principled approach to inferring user-perceived relevance of documents, which be leveraged in numerous applications businesses. Due huge volume data, scalability must.We present chain model (CCM), based on solid, Bayesian framework. both scalable...
Abstract Despite numerous assessments of the sensitivity and resilience drylands to degradation, there has been little research into way affected communities innovate adapt in response land degradation. This paper shows how local scientific knowledge can be combined identify rangeland management strategies reduce or To achieve this, we have developed applied a four‐stage social learning approach based on stakeholder participation three degradation ‘hotspots’ communal rangelands Kalahari,...
Abstract. We demonstrate improvements in CALIPSO (Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations) dust extinction retrievals over northern Africa Europe when corrections are applied regarding the Saharan lidar ratio assumption, separation of portion detected mixtures, averaging scheme introduced Level 3 product. First, a universal, spatially constant 58 sr instead 40 is to individual 2 dust-related backscatter products. The resulting aerosol optical depths show an...
Journal Article Singular Integrals and Elliptic Boundary Problems on Regular Semmes–Kenig–Toro Domains Get access Steve Hofmann, Hofmann 1Mathematics Department, University of Missouri, Columbia, MO 65211, USA Correspondence to be sent to: met@math.unc.edu Search for other works by this author on: Oxford Academic Google Scholar Marius Mitrea, Mitrea Michael Taylor 2Mathematics North Carolina, Chapel Hill, NC 27599, International Mathematics Research Notices, Volume 2010, Issue 14, Pages...
Abstract. This study assesses the impact of dust on surface solar radiation focussing an extreme event. For this purpose, we exploited synergy AERONET measurements and passive active satellite remote sensing (MODIS CALIPSO) observations, in conjunction with radiative transfer model (RTM) chemical transport (CTM) simulations 1-day forecasts from Copernicus Atmosphere Monitoring Service (CAMS). The area interest is eastern Mediterranean where anomalously high aerosol loads were recorded...
Methods for estimating the spatial distribution of PM2.5 concentrations have been developed but not yet able to effectively include correlation. We report on development a back-propagation neural network (S-BPNN) model designed specifically make such correlations implicit by incorporating lag variable (SLV) as virtual input variable. The S-BPNN fits nonlinear relationship between ground-based air quality monitoring station measurements PM2.5, satellite observations aerosol optical depth,...
We study the problem of determining a complete Riemannian manifold with boundary from Cauchy data harmonic functions.This arises in electrical impedance tomography, where one tries to find an unknown conductivity inside given body measurements done on body.Here, we show that can reconstruct complete, real-analytic, M compact set data, non-empty open subset Γ boundary, all functions Dirichlet supported Γ, provided dim ≥ 3. note for this result need no assumption topology other than...
A query independent feature, relating perhaps to document content, linkage or usage, can be transformed into a static, per-document relevance weight for use in ranking. The challenge is find good function transform feature values scores. This paper presents FLOE, simple density analysis method modelling the shape of transformation required, based on training data and without assuming independence between baseline. For new it addresses questions: required ranking, what sort appropriate and,...
Optimising the parameters of ranking functions with respect to standard IR rank-dependent cost has eluded satisfactory analytical treatment. We build on recent advances in alternative differentiable pairwise functions, and show that these techniques can be successfully applied tuning an existing family scoring (BM25), sense we cannot do better using sensible search heuristics directly optimize rank-based function NDCG. also demonstrate how size training set affects number hope tune this way.
This study estimates the impact of dust aerosols on surface solar radiation and energy in Egypt based Earth Observation (EO) related techniques. For this purpose, we exploited synergy monthly mean daily post processed satellite remote sensing observations from MODerate resolution Imaging Spectroradiometer (MODIS), radiative transfer model (RTM) simulations utilizing machine learning, conjunction with 1-day forecasts Copernicus Atmosphere Monitoring Service (CAMS). As cloudy conditions region...
Abstract. This study focuses on the assessment of surface solar radiation (SSR) based operational neural network (NN) and multi-regression function (MRF) modelling techniques that produce instantaneous (in less than 1 min) outputs. Using real-time cloud aerosol optical properties inputs from Spinning Enhanced Visible Infrared Imager (SEVIRI) board Meteosat Second Generation (MSG) satellite Copernicus Atmosphere Monitoring Service (CAMS), respectively, these models are capable calculating SSR...
Abstract Exposure biases are a pervasive non‐climatic change in land air temperature records which have been introduced as result of changes the way thermometers were sheltered from solar radiation and elements over time. not widely accounted for observational records, due to difficulties detecting/correcting bias using traditional homogenisation techniques; therefore, exposure still contribute significant uncertainty early period global compilations. Here, an empirical approach address...