Network representation learning based on community-aware and adaptive random walk for overlapping community detection

SiMAIM: identifying sockpuppets and puppetmasters on a single forum-oriented social media site

With the Internet becoming indispensable in our lives, social media has become an integral part of our lives. However, with this has come the phenomenon of a single user registering multiple accounts (sockpuppets) to advertise, spam, or cause controversy on social media sites, where the user is called the puppetmaster. This phenomenon is even more evident on forum-oriented social media sites. Identifying sockpuppets is a critical step in stopping the above-mentioned malicious acts. The identification of sockpuppets on a single forum-oriented social media site has seldom been addressed. This paper proposes a Single-site Multiple Accounts Identification Model (SiMAIM) framework to address this research gap. We used Mobile01, Taiwan's most popular forum-oriented social media site, to validate SiMAIM's performance. SiMAIM achieved F1 scores between 0.6 and 0.9 on identifying sockpuppets and puppetmasters under different datasets and settings. SiMAIM also outperformed the compared methods by 6-38% in F1 score.

A hybrid framework for delineating the migration route of soil heavy metal pollution by heavy metal similarity calculation and machine learning method

Soil heavy metal contamination was a global environmental issue that posed adverse impacts on ecological and human health risks. The controlling of soil heavy metal is mainly focused on the emission source and pipe-end treatment, less is known about the intermediate controlling process. The migration route of heavy metals exhibited the spatial evolution of pollutants from the sources to the pipe-end, which provided the more reasonable location for the target-oriented treatment of soil heavy metal. Here, we proposed a new view of heavy metal similarity, which quantitatively expressed how closely of the contaminations between the study area and the test areas. We found that the similarity of different heavy metals was unequally distributed across locations that were related with five main sources, namely agricultural activities, natural sources, traffic emissions, industrial activities, and other sources. Based on the similarity, a state-of-the-art machine learning method was applied to delineate the migration route of soil heavy metals. Thereinto, As was concentrated around livestock farms, and its migration route was close to the water system. Cd migration route was over-dispersed in the areas where located mine fields and chemical plants. Migration routes of Hg and Pb were along rivers, which were related to agricultural activities and natural sources. Overall, the perspective on similarity and migration routes provided theoretical basis and method to alleviate soil heavy metal pollution at regional scale and can be extended across largescale regions.

Discovering a cohesive football team through players’ attributed collaboration networks

The process of team composition in multiplayer sports such as football has been a main area of interest within the field of the science of teamwork, which is important for improving competition results and game experience. Recent algorithms for the football team composition problem take into account the skill proficiency of players but not the interactions between players that contribute to winning the championship. To automate the composition of a cohesive team, we consider the internal collaborations among football players. Specifically, we propose a Team Composition based on the Football Players’ Attributed Collaboration Network (TC-FPACN) model, aiming to identify a cohesive football team by maximizing football players’ capabilities and their collaborations via three network metrics, namely, network ability, network density and network heterogeneity&homogeneity. Solving the optimization problem is NP-hard; we develop an approximation method based on greedy algorithms and then improve the method through pruning strategies given a budget limit. We conduct experiments on two popular football simulation platforms. The experimental results show that our proposed approach can form effective teams that dominate others in the majority of simulated competitions.