Characteristics of Leakage Pollution of Longpan Road Gas Station and Its Enlightenment

Geological penetrating radar combined with drilling and chemical analysis has been applied to investigate leakage pollution of Longpan Road gas station in Nanjing, China. The results indicate that radar images show strong reflection anomalies along the northeast to the gas station, characterized by contaminants or contaminant plumes spreading downstream and below. The drilling results confirmed the contents of monocyclic and polycyclic aromatic hydrocarbons contained in the layers of fine sands ranging from 0.60 m to 6.0 m beneath the surface mostly exceed Chinese standard severely, such as toluene and isobutylbenzene with high content at 2738 μg/kg and 64505 μg/kg, respectively. Therefore, it is considered that geological penetrating radar can be employed to investigate leakage contamination of gas stations, and remediation and administration should be conducted in the polluted soil layers and aquifers.

Fact-condition statements and super relation extraction for geothermic knowledge graphs construction

Researchers utilize information from the geoscience literature to deduce the regional or global geological evolution.Traditionally this process has relied on the labor of researchers.As the number of papers continues to increase,acquiring domain-specific knowledge becomes a heavy burden.Knowledge Graph(KG)is proposed as a new knowledge representation technology to change this situation.However,the super relation is not considered in the previous KG,which bridges the geological phenomenon(fact)and its precondition(condition).For instance,in the statement(“the late Archean was a crucial transition period in the history of global geodynamics”),the condition statement(“crucial transition for global geodynamics”)works as the complementary fact statement(“the late Archean was a crucial transition period”),which defines the scale of crucial transition accurately in the late Archean.In this study,fact-condition statement extraction is introduced to construct a geological knowledge graph.A rule-based multi-input multi-output model(R-MIMO)is proposed for information extraction.In the R-MIMO,fact-condition statements and their super relation are considered and extracted for the first time.To verify its performances,a GeothCF dataset with 1455 fact tuples and 789 condition tuples is constructed.In experiments,the R-MIMO model achieves the best performance by using BERT as encoder and LSTM-d as decoder,achieving F180.24%in tuple extraction and F170.03%in tag prediction task.Furthermore,the geothermic KG with super relation is automatically constructed for the first time by trained R-MIMO,which can provide structured data for further geothermic research.

Region-aware neural graph collaborative filtering for personalized recommendation

Personalized recommender systems have been widely deployed in various scenarios to enhance user experience in response to the challenge of information explosion.Especially,personalized recommendation models based on graph structure have advanced greatly in predicting user preferences.However,geographical region entities that reflect the geographical context of the items is not being utilized in previous works,leaving room for the improvement of personalized recommendation.This study proposes a region-aware neural graph collaborative filtering(RA-NGCF)model,which introduces the geographical regions for improving the prediction of user preference.The approach first characterizes the relationships between items and users with a user-item-region graph.And,a neural network model for the region-aware graph is derived to capture the higher-order interaction among users,items,and regions.Finally,the model fuses region and item vectors to infer user preferences.Experiments on real-world dataset results show that introducing region entities improves the accuracy of personalized recommendations.This study provides a new approach for optimizing personalized recommendation as well as a methodological reference for facilitating geographical regions for optimizing spatial applications.

Combinatorial principles between RRT2 2 and RT2 2

We study the strength of some combinatorial principles weaker than Ramsey theorem for pairs over RCA0. First, we prove that Rainbow Ramsey theorem for pairs does not imply Thin Set theorem for pairs. Furthermore, we get some other related results on reverse mathematics using the same method. For instance, Rainbow Ramsey theorem for pairs is strictly weaker than ErdSs- Moser theorem under RCA0.