Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI

http://www.geoplanning.it/test/wp-content/uploads/2012/02/Practical-estimates-of-rock-mass-strength.pdf

Practical estimates of rock mass strength

本文对国际科学计量学杂志《Ｓｃｉｅｎｔｏｍｅｔｒｉｃｓ》１９７９－１９９１年的研究论文内容、栏目、作者及国别和编委及国别作了计量分析，揭示出科学计量学研究的重点、活动的中心及发展趋势，说明了学科带头人在发展科学计量学这门新兴学科中的作用。

国际刊物《Scientometrics》文献计量研究

In the last years, Artificial Intelligence (AI) has achieved a notable momentum that may deliver the best of expectations over many application sectors across the field. For this to occur, the entire community stands in front of the barrier of explainability, an inherent problem of AI techniques brought by sub-symbolism (e.g. ensembles or Deep Neural Networks) that were not present in the last hype of AI. Paradigms underlying this problem fall within the so-called eXplainable AI (XAI) field, which is acknowledged as a crucial feature for the practical deployment of AI models. This overview examines the existing literature in the field of XAI, including a prospect toward what is yet to be reached. We summarize previous efforts to define explainability in Machine Learning, establishing a novel definition that covers prior conceptual propositions with a major focus on the audience for which explainability is sought. We then propose and discuss about a taxonomy of recent contributions related to the explainability of different Machine Learning models, including those aimed at Deep Learning methods for which a second taxonomy is built. This literature analysis serves as the background for a series of challenges faced by XAI, such as the crossroads between data fusion and explainability. Our prospects lead toward the concept of Responsible Artificial Intelligence, namely, a methodology for the large-scale implementation of AI methods in real organizations with fairness, model explainability and accountability at its core. Our ultimate goal is to provide newcomers to XAI with a reference material in order to stimulate future research advances, but also to encourage experts and professionals from other disciplines to embrace the benefits of AI in their activity sectors, without any prior bias for its lack of interpretability.

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI.

/pdf/landslide-inventory-maps-new-tools-for-an-old-problem-wkxin2jqxg.pdf

Landslide inventory maps: New tools for an old problem

Abstract In the planning of the tunnel support system, besides the detailed investigation of the geological-geotechnical conditions, it is also important to investigate the structure-tunnel interaction. Especially the structures located in the close vicinity of the tunnels built in unpredictable geological conditions should be considered as a part of the tunnel support system design and the planning should be performed. However, the strong heterogeneity and anisotropy characteristics of karstic environments affect the planning. In this study, a single-track railway tunnel excavated in a relatively thin-bedded, karstic limestone in the Gebze Köseköy Railway project is investigated in terms of structure-tunnel interaction and the importance of grouting. The fact that the tunnel, located in the center of the two piers of the Osmangazi Suspension Bridge approach viaducts, built on the Gulf of Izmit, passes through a shallow overburden, is also an important difficulty in the design of the tunnel. Some karstic caves are determined in the limestones besides its heavily fractured nature. It is concluded that there is a possibility of damage to the bridge as a result of different settlements under the viaduct piers during the tunnel construction with the support to be performed only through the tunnel, and the karstic caves should be filled in order to prevent possible damages. The karstic limestones along the tunnel route are strengthened from the surface by injection of a mixture of water, cement, bentonite, and sand. The interaction of the tunnel with the viaduct after injection is investigated with 2D and 3D numerical analysis and tunnel excavation and support works are started. With the tunnel construction, measurements taken from inside the tunnel and from the viaduct piers, values very close to the predicted deformation limits were obtained by 3D numerical analysis, and the tunnel construction is successfully completed. Despite the extremely difficult conditions, the main reason why no problems are encountered during the tunnel construction is considered to be the grouting. 

/pdf/importance-of-grouting-for-tunneling-in-karstic-and-complex-22fb1e6v.pdf

Importance of grouting for tunneling in karstic and complex environment (a case study from Türkiye)

The main purposes of the present study were to compare in situ performances of two ground improvement techniques, which are jet grout and deep soil mixing techniques. In this study, Manisa State Hospital Project location was selected as the test site, because this location has difficult geological and geotechnical conditions. In the test site, three jet grout colons and three deep soil mix colons having same cement content were constructed. After 28 days of curing time, the core specimens were extracted for the uniaxial compressive testing, and lump specimens were collected for further tests and analyses such as X-ray diffraction, electron microscopy and determination of physical properties. The results of the uniaxial compression tests show that the uniaxial compressive strengths of the specimens extracted from the deep soil mix colons are drastically higher than those of the jet grout colons. Additionally, to understand the causes of this difference, several features of the improved soil were investigated by various mineralogical and petrographic methods. According to the results of these investigations, the fabric and some mineralogical properties were seriously affected by the improvement technique.

An Experimental Study to Compare Two Soil Improvement Techniques Performance

: Shallow landslides, which are generally triggered by extreme precipitation events, are increas-ingly becoming common in the world. Societies have had difficulty in keeping up with the exponentially rising rate of shallow landslides in recent years. Despite the considerable progress made in engineering studies, shallow landslides continue to cause considerable damage in different areas of the planet. Therefore, runout analyses are becoming more and more popular ways of building resilience to the negative effects of shallow landslides. Runout analyses are such crucial parts of shallow landslide studies that researchers have been keen to contribute to the existing knowledge on the subject. Earlier research suggested that runout analyses can be studied with empirical–statistical and numerical methods. Although there exist numerous landslide runout studies related to empirical–statistical and numerical solutions, we had not encountered a comparison of empirical–statistical and numerical methods’ advantages and disadvantages in the literature. This research presents an evaluation of the advantages and disadvantages of the runout analysis methods.

/pdf/a-review-of-the-prediction-methods-for-landslide-runout-1vtj24he.pdf

A Review of the Prediction Methods for Landslide Runout

Abstract Preparation of accurate and up-to-date susceptibility maps at the regional scale is mandatory for disaster mitigation, site selection, and planning in areas prone to multiple natural hazards. In this study, we proposed a novel multi-hazard susceptibility assessment approach that combines expert-based and supervised machine learning methods for landslide, flood, and earthquake hazard assessments for a basin in Elazig Province, Türkiye. To produce the landslide susceptibility map, an ensemble machine learning algorithm, random forest, was chosen because of its known performance in similar studies. The modified analytical hierarchical process method was used to produce the flood susceptibility map by using factor scores that were defined specifically for the area in the study. The seismic hazard was assessed using ground motion parameters based on Arias intensity values. The univariate maps were synthesized with a Mamdani fuzzy inference system using membership functions designated by expert. The results show that the random forest provided an overall accuracy of 92.3% for landslide susceptibility mapping. Of the study area, 41.24% were found prone to multi-hazards (probability value &gt; 50%), but the southern parts of the study area are more susceptible. The proposed model is suitable for multi-hazard susceptibility assessment at a regional scale although expert intervention may be required for optimizing the algorithms. 

/pdf/a-hybrid-multi-hazard-susceptibility-assessment-model-for-a-3pdc6ya5.pdf

A Hybrid Multi-Hazard Susceptibility Assessment Model for a Basin in Elazig Province, Türkiye

When seismic and multi-beam bathymetric data from the northern shelf and slope of the Cinarcik Basin are interpreted, some sub-marine landslides are observed clearly. Additionally, seismic data indicate that upper surface of the submarine extension of the Paleozoic aged rocks has NNE-SSW oriented basin and ridge type morphology controlled by the secondary faults of the NAFZ. Basins are fulfilled by Plio-Quaternary sediments, which are cut by strike-slif faults on the shelf and slope. Thickness of these deposits increases up to 130 m toward the concave shaped northern slope of the Cinarcik Basin. A relatively recent sub-marine landslide, Tuzla Sub-marine Landslide, cuts the concave slope of the Cinarcik Basin. Detailed morphological investigation indicate that Tuzla Landslide is a deep-seated rotational landslide, which possibly triggered by the NAFZ. Morphological analyses also indicate that thick Plio-Quaternary deposits on the Paleozoic Basement were slided during the Tuzla Landslide event. This landslide is considered as a key event for modeling the future landslide potential of the northern shelf and slope of the Cinarcik Basin. For this reason, the main purpose of the present study is to perform some rheological analyses to understand the behaviour of the events. As the main results obtained from the analyses, the runout distances and the velocities were calculated.

https://geolib.geo.auth.gr/index.php/sasg/article/download/7591/7348

Candan Gokceoglu

Papers

Importance of grouting for tunneling in karstic and complex environment (a case study from Türkiye)

An Experimental Study to Compare Two Soil Improvement Techniques Performance

A Review of the Prediction Methods for Landslide Runout

A Hybrid Multi-Hazard Susceptibility Assessment Model for a Basin in Elazig Province, Türkiye

Rheological analysis of a sub-marine landslide in the Marmara sea (Turkey)