Searcharxiv⌕ Search

arXiv subjects

Radosław Michalski

Publications and source records attributed to Radosław Michalski.

25 records · Page 2Linked to original sources

Learning in Unlabeled Networks - An Active Learning and Inference Approach

The task of determining labels of all network nodes based on the knowledge about network structure and labels of some training subset of nodes is called the within-network classification. It may happen that none of the labels of the nodes is known and additionally there is no information about number of classes to which nodes can be assigned. In such a case a subset of nodes has to be selected for initial label acquisition. The question that arises is: "labels of which nodes should be collected and used for learning in order to provide the best classification accuracy for the whole network?". Active learning and inference is a practical framework to study this problem. A set of methods for active learning and inference for within network classification is proposed and validated. The utility score calculation for each node based on network structure is the first step in the process. The scores enable to rank the nodes. Based on the ranking, a set of nodes, for which the labels are acquired, is selected (e.g. by taking top or bottom N from the ranking). The new measure-neighbour methods proposed in the paper suggest not obtaining labels of nodes from the ranking but rather acquiring labels of their neighbours. The paper examines 29 distinct formulations of utility score and selection methods reporting their impact on the results of two collective classification algorithms: Iterative Classification Algorithm and Loopy Belief Propagation. We advocate that the accuracy of presented methods depends on the structural properties of the examined network. We claim that measure-neighbour methods will work better than the regular methods for networks with higher clustering coefficient and worse than regular methods for networks with low clustering coefficient. According to our hypothesis, based on clustering coefficient we are able to recommend appropriate active learning and inference method.

stat.ML↗

Knowledge Acquisition from Social Platforms Based on Network Distributions Fitting

The uniqueness of online social networks makes it possible to implement new methods that increase the quality and effectiveness of research processes. While surveys are one of the most important tools for research, the representativeness of selected online samples is often a challenge and the results are hardly generalizable. An approach based on surveys with representativeness targeted at network measure distributions is proposed and analysed in this paper. Its main goal is to focus not only on sample representativeness in terms of demographic attributes, but also to follow the measures distributions within main network. The approach presented has many application areas related to online research, sampling a network for the evaluation of collaborative learning processes, and candidate selection for training purposes with the ability to distribute information within a social network.

cs.SI↗

Seed Selection for Spread of Influence in Social Networks: Temporal vs. Static Approach

The problem of finding optimal set of users for influencing others in the social network has been widely studied. Because it is NP-hard, some heuristics were proposed to find sub-optimal solutions. Still, one of the commonly used assumption is the one that seeds are chosen on the static network, not the dynamic one. This static approach is in fact far from the real-world networks, where new nodes may appear and old ones dynamically disappear in course of time. The main purpose of this paper is to analyse how the results of one of the typical models for spread of influence - linear threshold - differ depending on the strategy of building the social network used later for choosing seeds. To show the impact of network creation strategy on the final number of influenced nodes - outcome of spread of influence, the results for three approaches were studied: one static and two temporal with different granularities, i.e. various number of time windows. Social networks for each time window encapsulated dynamic changes in the network structure. Calculation of various node structural measures like degree or betweenness respected these changes by means of forgetting mechanism - more recent data had greater influence on node measure values. These measures were, in turn, used for node ranking and their selection for seeding. All concepts were applied to experimental verification on five real datasets. The results revealed that temporal approach is always better than static and the higher granularity in the temporal social network while seeding, the more finally influenced nodes. Additionally, outdegree measure with exponential forgetting typically outperformed other time-dependent structural measures, if used for seed candidate ranking.

cs.SI↗

Quantifying Social Network Dynamics

The dynamic character of most social networks requires to model evolution of networks in order to enable complex analysis of theirs dynamics. The following paper focuses on the definition of differences between network snapshots by means of Graph Differential Tuple. These differences enable to calculate the diverse distance measures as well as to investigate the speed of changes. Four separate measures are suggested in the paper with experimental study on real social network data.

cs.SI↗

The Multidimensional Study of Viral Campaigns as Branching Processes

Viral campaigns on the Internet may follow variety of models, depending on the content, incentives, personal attitudes of sender and recipient to the content and other factors. Due to the fact that the knowledge of the campaign specifics is essential for the campaign managers, researchers are constantly evaluating models and real-world data. The goal of this article is to present the new knowledge obtained from studying two viral campaigns that took place in a virtual world which followed the branching process. The results show that it is possible to reduce the time needed to estimate the model parameters of the campaign and, moreover, some important aspects of time-generations relationship are presented.

cs.SI↗

Studying Paths of Participation in Viral Diffusion Process

Authors propose a conceptual model of participation in viral diffusion process composed of four stages: awareness, infection, engagement and action. To verify the model it has been applied and studied in the virtual social chat environment settings. The study investigates the behavioral paths of actions that reflect the stages of participation in the diffusion and presents shortcuts, that lead to the final action, i.e. the attendance in a virtual event. The results show that the participation in each stage of the process increases the probability of reaching the final action. Nevertheless, the majority of users involved in the virtual event did not go through each stage of the process but followed the shortcuts. That suggests that the viral diffusion process is not necessarily a linear sequence of human actions but rather a dynamic system.

cs.SI↗

Negative Effects of Incentivised Viral Campaigns for Activity in Social Networks

Viral campaigns are crucial methods for word-of-mouth marketing in social communities. The goal of these campaigns is to encourage people for activity. The problem of incentivised and non-incentivised campaigns is studied in the paper. Based on the data collected within the real social networking site both approaches were compared. The experimental results revealed that a highly motivated campaign not necessarily provides better results due to overlapping effect. Additional studies have shown that the behaviour of individual community members in the campaign based on their service profile can be predicted but the classification accuracy may be limited.

cs.SI↗