The Adaptive Boosting (AdaBoost) algorithm is a widely used ensemble learning framework, and it can get good classification results on general datasets. However, it is challenging to apply the AdaBoost algorithm directly to imbalanced data since it is designed mainly for processing misclassified samples rather than samples of minority classes. To better process imbalanced data, this paper introduces the indicator Area Under Curve (AUC) which can reflect the comprehensive performance of the model, and proposes an improved AdaBoost algorithm based on AUC (AdaBoost-A) which improves the error calculation performance of the AdaBoost algorithm by comprehensively considering the effects of misclassification probability and AUC. To prevent redundant or useless weak classifiers the traditional AdaBoost algorithm generated from consuming too much system resources, this paper proposes an ensemble algorithm, PSOPD-AdaBoost-A, which can re-initialize parameters to avoid falling into local optimum, and optimize the coefficients of AdaBoost weak classifiers. Experiment results show that the proposed algorithm is effective for processing imbalanced data, especially the data with relatively high imbalances.

Joint seismic-electromagnetic surveys are increasingly being used for understanding the subsurface. Valid rock-physics models are key for interpreting joint survey data. Such models should explore the cross-property relationship between the elastic and electrical properties and link the joint elastic-electrical correlations to the properties of various types of rocks of interest. However, although current differential effective medium (DEM) models have had great success in independently modeling the elastic and electrical properties of porous rock, they are not directly suitable for joint elastic-electrical modeling because the elastic and electrical DEM models describe the microstructures of the rocks in different ways. Based on dedicated laboratory measurements and theoretical modeling, we have investigated how the effective pore and grain shapes are correlated in two Berea sandstones. We determine that the effective pore aspect ratios inverted from the elastic properties exhibit a negative linear correlation with the effective grain aspect ratios obtained from the electrical inversion. The effective pore and grain aspect ratios are linked as an implicit function of varying porosity with confining pressure. Using the derived pore and grain shapes correlation, the joint elastic-electrical modeling results compared successfully with experimental data. The obtained results reveal the natural link between the effective pore and grain shapes in Berea sandstones and make the microstructures between the elastic and electrical models consistent.

The Ri-Qing-Wei Basin is a newly discovered Late Mesozoic rift basin on the eastern Shandong coast in recent years. Thick, continuous deposited source rocks are found in the Lingshan Island scientific drilling project. Therefore, it is necessary to evaluate the distribution and hydrocarbon generation potential of source rocks in this area. Organic geochemical experiments were carried out on samples from the core of LK-1 and outcrops in the Lingshan Island, the Laoshan area, and the Jimo Zhougezhuang area to evaluate the maturity, abundance, and type of organic matter in source rocks. The results show that the cumulative thickness of the source rocks in the study area is more than 500 m, and the TOC content is generally greater than 1.0%. The organic matter type is good (mainly type II₁ and type II₂ kerogen) and the R_o value is more than 2.0. The thermal evolution degree of the organic matter is high with natural gas predominantly generated. We carried out Rock-Eval, PY-GC and gold tube experiments on low-maturity samples of Laikong 2 and established a chemical kinetic model to quantitatively evaluate the study area in combination with sedimentary burial history and thermal history data. The results show that the total resources of the five sags in the study area are about 476 billion cubic meters, and the average resource intensity is about 82.2 million cubic meters/km². Among them, the resource intensities of the Lingshan Island Sag and Laoshan Sag are 112.6 million cubic meters/km² and 98.8 million cubic meters/km², respectively. Studied sites are “small and fertile” and may be used as favorable exploration prospect areas.