Semi-supervised multi-class Adaboost by exploiting unlabeled data

Semi-supervised learning has attracted much attention in pattern recognition and machine learning. Most semi-supervised learning algorithms are proposed for binary classification, and then extended to multi-class cases by using approaches such as one-against-the-rest. In this work, we propose a semi-supervised learning method by using the multi-class boosting, which can directly classify the multi-class data and achieve high classification accuracy by exploiting the unlabeled data. There are two distinct features in our proposed semi-supervised learning approach: (1) handling multi-class cases directly without reducing them to multiple two-class problems, and (2) the classification accuracy of each base classifier requiring only at least 1/K or better than 1/K (K is the number of classes). Experimental results show that the proposed method is effective based on the testing of 21 UCI benchmark data sets.     

SFFS-PC-NN optimized by genetic algorithm for dynamic prediction of financial distress with longitudinal data streams

Recently, research of financial distress prediction has become increasingly urgent. However, existing static models for financial distress prediction are not able to adapt to the situation that the sample data flows constantly with the lapse of time. Financial distress prediction with static models does not meet the demand of the dynamic nature of business operations. This article explores the theoretical and empirical research of dynamic modeling on financial distress prediction with longitudinal data streams from the view of individual enterprise. Based on enterprise’s longitudinal data streams, dynamic financial distress prediction model is constructed by integrating financial indicator selection by using sequential floating forward selection method, dynamic evaluation of enterprise’s financial situation by using principal component analysis at each longitudinal time point, and dynamic prediction of financial distress by using back-propagation neural network optimized by genetic algorithm. This model’s ex-ante prediction efficiently combines its ex-post evaluation. In empirical study, three listed companies’ half-year longitudinal data streams are used as the sample set. Results of dynamic financial distress prediction show that the longitudinal and dynamic model of enterprise’s financial distress prediction is more effective and feasible than static model.     

First genetically verified occurrence of Ligula pavlovskii outside its native range and characteristics of its infection in Neogobius fluviatilis

In this study, we provide the first genetically verified distribution record beyond its native range of Ligula pavlovskii, a high-impact endoparasite of Ponto-Caspian gobies. According to parasitological surveys, ligulosis was detected for the first time in monkey goby individuals collected from Lake Balaton in 2004, 34 years after the first record of monkey goby in the lake. During a Lake Balaton survey in 2018, we detected tapeworms in 44.4% of the sampled monkey gobies. This prevalence is about two to ten times higher than it was reported from within the native range of the fish hosts. The same survey revealed an uneven spatial distribution of the ligulosis in the monkey gobies inhabiting different shoreline sections of the lake. We assume that the occurrence and recent distribution data of this high-impact endoparasite may provide useful information for potential biocontrol measures of invasive Ponto-Caspian gobies in the future.     

国内跨界水污染治理研究综述

跨界水污染是我国最严重的环境问题之一,自20世纪90年代以来已有许多研究。这些研究成果主要涉及了两方面:一是跨界水污染治理问题的宏观制度分析;二是解决跨界水污染问题微观技术手段研究。本文阐述了现阶段国内关于跨界水污染问题治理的研究成果,并在此基础之上进一步探讨了跨界水污染治理研究中存在的不足,指出有待下一步研究的问题。     

Requirements for Autonomous Underwater Vehicles (AUVs) for scientific data collection in the Laurentian Great Lakes: A questionnaire survey

Using mobile environmental monitoring can aid in gathering ecological data to meet fish community goals in the Great Lakes. One such approach is the use of large Autonomous Underwater Vehicles (AUVs) to gather data, or the potential use of AUV swarms, where multiple small AUVs work together with each having different data-gathering capabilities. To understand data needs that could be collected by mobile sensor networks to inform decision making, we surveyed Great Lakes professionals involved directly and indirectly in such decision making. Basic data that respondents chose as most important to collect were water temperature, dissolved oxygen, chlorophyll a, turbidity, and blue-green “algae”, which seems to align with variables affecting fish directly or indirectly (through identification of harmful algal blooms). Specialized data chosen as most important were mapping of habitat characteristics, sonar of groupings of fish, and images/video. The time of year to collect all data was chosen as all seasons by the majority of respondents, the frequency most chosen was once a season for mapping of habitat characteristics, once a week for sonar detection of groupings of fish, and once per day for images/video and water temperature. Results were very similar when respondents were asked where data should be collected in the Great Lakes (i.e., tributaries, nearshore areas, etc.) except respondents indicated that images/video should be collected most in fish spawning habitats. Understanding data important to inform decisions of resource professionals will help guide the design of mobile and stationary sensor networks in the Great Lakes.     

Social network site use,mobile personal talk and social capital among teenagers

Social network sites (SNSs) and mobile phones are becoming increasingly important in teenagers’ lives. Using data collected from a nationally representative survey (N = 800), this study explores the variation of social capital by SNS adoption, different SNS activities, and mobile personal talk among teenagers. The results indicate that SNS adoption and mobile personal talk can not only enhance teenagers’ close ties with friends, but also jointly promote teenagers’ civic engagement. Among SNS users, mobile personal talk also increase teens’ network capital. Different SNS activities such as commenting on friend’s Facebook pictures and joining Facebook groups have different relationships with social capital, and such relationships are moderated by mobile personal talk.     

Self-regulation and social media addiction: A multi-wave data analysis in China

Based on social cognitive theory (SCT), the present study aimed to investigate the impact of social media addiction (SMA) on employee strain and task performance through work-technology conflict and self-esteem. Furthermore, this study hypothesizes that the moderating influence of self-regulation may buffer the impact of SMA on work-technology conflict and self-esteem. The final sample of this study was 282 employees of Chinese information technology companies from southwest China's Guizhou province. Results demonstrated that SMA significantly affects task performance via work-technology conflict and self-esteem. This study presents some distinctive results by showing that the mediating role of self-esteem is found to be insignificant in the association between SMA and strain. Self-regulation reduces the increased impact of SMA on work-technology conflict and self-esteem. Moreover, this study provides useful insights for practitioners, managers, and future researchers to understand the impact of SMA on employees' mental health and task performance.