In previous link prediction researches, most scholars evaluate the influence of endpoints by the degree or H-index of endpoints, resulting in limited prediction accuracy. Through abundant investigations, we can evaluate the influence of endpoints accurately by the hybrid influence of neighbor nodes. Meanwhile, we calculate the hybrid influence of neighbors (HIN) by the average values of degree and H-index. In the paper, we conceive a HIN model. Large-scale experiments on 12 real datasets indicate that the conceived methods can significantly enhance the accuracy of link prediction.
The compressibility of abnormal pressure gas reservoirs is hard to test, and the interpretation is confusing, leading to many misunderstandings in the current understanding of abnormal pressure gas reservoirs. In this research, a highpressure experimental system was designed, and a series of highpressure compressibility tests of pure water, nitrogen, and rocks under different water saturations were carried out. Then, the effective compressibility of gas reservoirs was calculated; the effect of water saturation on abnormal pressure gas reservoirs and the dynamic prediction was studied. The results show that the compressibilities of water and rock are effectively constant values over the range examined, while the compressibility of gas decreases exponentially with the increase in pressure. The effective compressibility of the stratum increases with the rise of water saturation. The theory of stress and strain of rock mechanics also shows that the rock compressibility is determined by Young's modulus, Poisson's ratio, and porosity and has no connection with the formation pressure. With the increase in water saturation, the swelling degree of the production indicator curve of the simulation experiment becomes larger and larger. After introducing the effective compressibility of the stratum into the gas−water material balance equation, the gas reserves predicted by the revised production indicator curve are the same as the original reserves. The research results have important guiding significance for the efficient development of gas reservoirs.
Performance improvement of topological similarity-based link prediction models becomes an important research in complex networks. In the models based on node influence, researchers mainly consider the roles of endpoints or neighbors. Through investigations, we find that an endpoint with large influence has many neighbors. Meanwhile, the neighbors connect with more nodes besides endpoint, meaning that the endpoint can transmit extensive influence by the powerful combination of itself and neighbors. In addition, we evaluate the node influence by degree because the degree represents the number of neighbors accurately. In this paper, through focusing on the degree of endpoints and neighbors, we propose the powerful combination of endpoints and neighbors (PCEN) model. Experiments on twelve real network datasets demonstrate that the proposed model has better prediction performances than the traditional 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.