Generic metadata representation framework for social-based event detection, description, and linkage

Abebe, Minale Ashagrie; Tekli, Joe; Getahun, Fekade; Chbeir, Richard; Tekli, Gilbert

doi:10.1016/j.knosys.2019.06.025

Cited by 34 publications

(21 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Haversine formula was used to calculate the distance D ud between the donor and ungauged catchments' outlets, the same as Qi et al (2021a). The Haversine formula determines the great-circle distance between two points on a sphere given their longitudes and latitudes (Abebe et al 2020). The D ud is calculated by Equation ( 4):…”

Section: Regionalization Methodsmentioning

confidence: 99%

Regionalization of catchment hydrological model parameters for global water resources simulations

Chen

et al. 2022

Hydrology Research

View full text Add to dashboard Cite

Parameter regionalization of hydrological models is one of the most commonly used methods for hydrological prediction over ungauged catchments. Although there were many regional studies, there is no clear conclusion on the best-performed regionalization method for global hydrological modelling. The objective of this study is to determine an appropriate global-scale regionalization scheme (GSRS) for global hydrological modelling. To this end, the performance of five regionalization methods with two different average options, two weighting approaches, and seven efficiency thresholds (i.e. Kling-Gupta efficiency (KGE) values to measure hydrological model performances) was compared over thousands of catchments based on four conceptual hydrological models. Results of nine global models from the Global Earth Observation for Integrated Water Resource Assessment (EartH2Observe) project were selected to validate the accuracy of GSRS in estimating global runoff. The results show that: (1) Spatial proximity method with the Inverse Distance Weighting method and the output average option offers the best regionalization result when using the KGE ≥ 0.5 as an efficiency threshold for all four hydrological models, (2) the regionalization-based global hydrological simulation schemes (RGHSs), i.e. the proposed GSRS combining with four hydrological models, consistently performs better than the nine global models from EartH2Observe project in the estimation of runoff for most catchments, with varying degrees of improvement in the median, upper and lower quartiles, and whiskers of each performance metric, and (3) the global long-term annual water resources estimated by RGHSs range between 42,592 and 46,810 km3/yr.

show abstract

Section: Regionalization Methodsmentioning

confidence: 99%

Regionalization of catchment hydrological model parameters for global water resources simulations

Chen

et al. 2022

Hydrology Research

View full text Add to dashboard Cite

show abstract

“…They obtained heterogeneous walking sequences by heterogeneous random walk, which can mine the relation of graph structure and triplet sample and outperforms low-rank representation. Abebe M A et al [101] introduced a generic Social-based Event Detection framework (SEDDaL), the model input a collection of social media objects from heterogeneous sources and produced semantically meaningful events interconnected with spatial, temporal, and semantic relationships.…”

Section: Deep Learning Based Methodsmentioning

confidence: 99%

“…Feature fusion based methods [79] Utilize various features Lack of social network information [80] Constrained clustering algorithm is used to achieved high accuracy Imbalanced data and parameter setting is not optimal [81] Automatic concept mining and boosted concept learning Application is limited [82] Unsupervised and without predefined threshold Lack of more types of semantic features [83] Need no manual annotation and can adapt concepts to news domains - [84] Utilized the information contained in the related exemplars -Matrix factorization based methods [85] Robust to data incompleteness - [86] Can discover the shared structure between the datasets - [90] semi-supervised co-clustering with side information Parameters setting is not automatic [91] Contain dictionary atoms that are semantically discriminative - [92] Predict image clicks and solved the problem of lack of data -Topic model based methods [94] Generate visualized summaries Lack of personalized microblog summarization [96] exploit the multimodality and suitable for large-scale data Without videos and audios modality [97] Exploit various property jointly and classify multi-class events - [98] Can classify the visual-representative topics from non-visual-representative topics Didn't consider different domains Deep learning based methods [99] restrains the negative influence of noisy or irrelevant concepts - [100] Maintain multi-view information by robust representation - [101] A generic model to describe events and their relationships -Other Methods [102] Deal with multi-view tasks Poor data quality and high complexity [103] Jointly regularizing the encoded representations Lack of event summarization and itemization [104] Retrieval structured event representation, robust and effective - [105] Structured topic representation -…”

Section: Reference Advantages Disadvantagesmentioning

confidence: 99%

A Survey of Data Representation for Multi-Modality Event Detection and Evolution

2022

View full text Add to dashboard Cite

The rapid growth of online data has made it very convenient for people to obtain information. However, it also leads to the problem of “information overload”. Therefore, how to detect hot events from the massive amount of information has always been a problem. With the development of multimedia platforms, event detection has gradually developed from traditional single modality detection to multi-modality detection and is receiving increasing attention. The goal of multi-modality event detection is to discover events from a huge amount of online data with different data structures, such as texts, images and videos. These data represent real-world events from different perspectives so that they can provide more information about an event. In addition, event evolution is also a meaningful research direction; it models how events change dynamically over time and has great significance for event analysis. This paper comprehensively reviews the existing research on event detection and evolution. We first give a series of necessary definitions of event detection and evolution. Next, we discuss the techniques of data representation for event detection, including textual, visual, and multi-modality content. Finally, we review event evolution under multi-modality data. Furthermore, we review several public datasets and compare their results. At the end of this paper, we provide a conclusion and discuss future work.

show abstract

“…For example, Yu et al [20] investigated event detection to support decisions. Sahoh and Choksuriwong [26] and Abebe et al [27] employed a semantics-aware event-based approach and discussed how critical thinking could be utilized in intelligent high-stakes systems. Xu et al [28] proposed heuristic-based event descriptions using critical thinking for detection.…”

Section: Critical Thinkingmentioning

confidence: 99%

Causal Artificial Intelligence for High-Stakes Decisions: The Design and Development of a Causal Machine Learning Model

2022

View full text Add to dashboard Cite

show abstract

Generic metadata representation framework for social-based event detection, description, and linkage

Cited by 34 publications

References 34 publications

Regionalization of catchment hydrological model parameters for global water resources simulations

Regionalization of catchment hydrological model parameters for global water resources simulations

A Survey of Data Representation for Multi-Modality Event Detection and Evolution

Causal Artificial Intelligence for High-Stakes Decisions: The Design and Development of a Causal Machine Learning Model

Contact Info

Product

Resources

About