Person-job fit is to match candidates and job posts on online recruitment platforms using machine learning algorithms. The effectiveness of matching algorithms heavily depends on the learned representations for the candidates and job posts. In this paper, we propose to learn comprehensive and effective representations of the candidates and job posts via feature fusion. First, in addition to applying deep learning models for processing the free text in resumes and job posts, which is adopted by existing methods, we extract semantic entities from the whole resume (and job post) and then learn features for them. By fusing the features from the free text and the entities, we get a comprehensive representation for the information explicitly stated in the resume and job post. Second, however, some information of a candidate or a job may not be explicitly captured in the resume or job post. Nonetheless, the historical applications including accepted and rejected cases can reveal some implicit intentions of the candidates or recruiters. Therefore, we propose to learn the representations of implicit intentions by processing the historical applications using LSTM. Last, by fusing the representations for the explicit and implicit intentions, we get a more comprehensive and effective representation for person-job fit. Experiments over 10 months real data show that our solution outperforms existing methods with a large margin. Ablation studies confirm the contribution of each component of the fused representation. The extracted semantic entities help interpret the matching results in the case study.
Leaf area index (LAI) is an important vegetation leaf structure parameter in forest and agricultural ecosystems. Remote sensing techniques can provide an effective alternative to field-based observation of LAI. Differences in canopy structure result in different sensor types (active or passive), platforms (terrestrial, airborne, or satellite), and models being appropriate for the LAI estimation of forest and agricultural systems. This study reviews the application of remote sensing-based approaches across different system configurations (passive, active, and multisource sensors on different collection platforms) that are used to estimate forest and crop LAI and explores uncertainty analysis in LAI estimation. A comparison of the difference in LAI estimation for forest and agricultural applications given the different structure of these ecosystems is presented, particularly as this relates to spatial scale. The ease of use of empirical models supports these as the preferred choice for forest and crop LAI estimation. However, performance variation among different empirical models for forest and crop LAI estimation limits the broad application of specific models. The development of models that facilitate the strategic incorporation of local physiology and biochemistry parameters for specific forests and crop growth stages from various temperature zones could improve the accuracy of LAI estimation models and help develop models that can be applied more broadly. In terms of scale issues, both spectral and spatial scales impact the estimation of LAI. Exploration of the quantitative relationship between scales of data from different sensors could help forest and crop managers more appropriately and effectively apply different data sources. Uncertainty coming from various sources results in reduced accuracy in estimating LAI. While Bayesian approaches have proven effective to quantify LAI estimation uncertainty based on the uncertainty of model inputs, there is still a need to quantify uncertainty from remote sensing data source, ground measurements and related environmental factors to mitigate the impacts of model uncertainty and improve LAI estimation.
Signals of Opportunity Reflectometry (SoOp-R) employs the communication system, GNSS (Global Navigation Satellite System) constellation and other potential Signals of Opportunity (SoOp) as the transmitters. In recent years, it has gained increased interests. Several experiments have been carried out, however it is still in the initial development stage. Theoretical predictions of SoOp Reflectometry for land surface parameters detection, such as soil moisture and vegetation biomass, should be carried out simultaneously. Meanwhile, at present less works are paid attention to the polarization study of the polarizations. The first-order radiative transfer equation models are employed here and they are developed according to the wave synthesis technique to get the various polarization combinations. Using the two models as analysis tools, we simulate the bistatic scattering at all potential SoOp Reflectometry bands, i.e., P-, L-, C- and X-band for circular polarizations and linear polarizations. While the original commonly used microwave scattering models are linear polarizations, here we compare the difference. Although the models can simulate bistatic scattering at any incident angles and scattering angles. Four special observation geometry are taken into considerations during the analysis. Using the developed models as tools, the developed models establish the relationship between the land surface parameters (such as soil moisture, soil roughness and vegetation water content, diameters et al.) and bistatic radar cross section. The forward scattering models developed here enables the understanding of the effects of different geophysical parameters and transmitter–receiver observation scenarios on the bisatic scattering at any polarization combinations for any potential SoOP reflectometry bands. Robust retrieval methods for soil moisture and vegetation biomass can benefit from the forward scattering models.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.