scholarly journals A Variational Point Process Model for Social Event Sequences

2020 ◽  
Vol 34 (01) ◽  
pp. 173-180
Author(s):  
Zhen Pan ◽  
Zhenya Huang ◽  
Defu Lian ◽  
Enhong Chen
Keyword(s):  
Point Process ◽  
Real World ◽  
Process Model ◽  
Classification Model ◽  
Event Sequence ◽  
Event Sequences ◽  
Sequence Modeling ◽  
Stochastic Point Process ◽  
Feature Based ◽  
Real World Datasets

Many events occur in real-world and social networks. Events are related to the past and there are patterns in the evolution of event sequences. Understanding the patterns can help us better predict the type and arriving time of the next event. In the literature, both feature-based approaches and generative approaches are utilized to model the event sequence. Feature-based approaches extract a variety of features, and train a regression or classification model to make a prediction. Yet, their performance is dependent on the experience-based feature exaction. Generative approaches usually assume the evolution of events follow a stochastic point process (e.g., Poisson process or its complexer variants). However, the true distribution of events is never known and the performance depends on the design of stochastic process in practice. To solve the above challenges, in this paper, we present a novel probabilistic generative model for event sequences. The model is termed Variational Event Point Process (VEPP). Our model introduces variational auto-encoder to event sequence modeling that can better use the latent information and capture the distribution over inter-arrival time and types of event sequences. Experiments on real-world datasets prove effectiveness of our proposed model.

scholarly journals A Neural Approach for Modeling Continuous Time Sequences with Intermittent Observations

2021 ◽  
Author(s):  
Vinayak Gupta ◽  
Srikanta Bedathur
Keyword(s):  
Real World ◽  
Continuous Time ◽  
Point Processes ◽  
Large Fraction ◽  
Modeling Framework ◽  
Event Sequences ◽  
Event Time ◽  
Sequence Modeling ◽  
Recent Success ◽  
Real World Datasets

A large fraction of data generated via human activities such as online purchases, health records, spatial mobility etc. can be represented as continuous-time event sequences (CTES) i.e. sequences of discrete events over a continuous time. Learning neural models over CTES is a non-trivial task as it involves modeling the ever-increasing event timestamps, inter-event time gaps, event types, and the influences between different events within and across different sequences. Moreover, existing sequence modeling techniques consider a complete observation scenario i.e. the event sequence being modeled is completely observed with no missing events – an ideal setting that is rarely applicable in real-world applications. In this paper, we highlight our approach[8] for modeling CTES with intermittent observations. Buoyed by the recent success of neural marked temporal point processes (MTPP) for modeling the generative distribution of CTES, we provide a novel unsupervised model and inference method for learning MTPP in presence of event sequences with missing events. Specifically, we first model the generative processes of observed events and missing events using two MTPP, where the missing events are represented as latent random variables. Then, we devise an unsupervised training method that jointly learns both the MTPP using variational inference. Experiments across real-world datasets show that our modeling framework outperforms state-of-the-art techniques for future event prediction and imputation. This work appeared in AISTATS 2021.

Effect of Type-II and Type-III PMs on Operational Reliability of Buses

2010 ◽  
Vol 118-120 ◽  
pp. 309-313 ◽  
Author(s):  
R. Jiang ◽  
Guo Feng Zhang
Keyword(s):  
Point Process ◽  
Power Law ◽  
Real World ◽  
Preventive Maintenance ◽  
Process Model ◽  
Operational Reliability ◽  
Type Ii ◽  
Real World Data ◽  
Type Iii ◽  
Failure Point

This paper deals with modeling the operational reliability of a bus fleet. The operational reliability is represented by the rate of occurrence of minor failures. A power-law failure point process model with the cumulative numbers of type-II and type-III preventive maintenance (PM) as covariates is fitted to a set of real-world data. We evaluate the effect of these two types of PM on minor failures of buses using the fitted model. This provides a method to quantitatively evaluate the effectiveness of a kind of PM actions.

scholarly journals A General Two-stage Multi-label Ranking Framework

2021 ◽  
Vol 34 (1) ◽  
Author(s):  
Yanbing Xue ◽  
Milos Hauskrecht
Keyword(s):  
Real World ◽  
Classification Model ◽  
Stage Model ◽  
Two Stage ◽  
Label Ranking ◽  
Improved Performance ◽  
Real World Datasets ◽  
Post Hoc ◽  
Do So

In this paper we develop and study solutions for the multi-label ranking (MLR) problem. Briefly, the goal of multi-label ranking is not only to assign a set of relevant labels to a data instance but also to rank the labels according to their importance. To do so we propose a two-stage model that consists of: (1) a multi-label classification model that first selects an unordered set of labels for a data instance, and, (2) a label ordering model that orders the selected labels post-hoc in order of their importance. The advantage of such a model is that it can represent both the dependencies among labels, as well as, their importance. We evaluate the performance of our framework on both simulated and real-world datasets and show its improved performance compared to the existing multiple-label ranking solutions.

scholarly journals Determinants of spatial intensity of stop locations on cruise passengers tracking data

2021 ◽  
pp. 159-164
Author(s):  
Nicoletta D’Angelo ◽  
Mauro Ferrante ◽  
Antonino Abbruzzo ◽  
Giada Adelfio
Keyword(s):  
Point Process ◽  
Process Model ◽  
Linear Network ◽  
Spatial Intensity ◽  
Quadrature Scheme ◽  
Stochastic Point Process ◽  
Gibbs Point Process ◽  
Location Patterns ◽  
Level Information ◽  
Main Determinants

This paper aims at analyzing the spatial intensity in the distribution of stop locations of cruise passengers during their visit at the destination through a stochastic point process modelling approach on a linear network. Data collected through the integration of GPS tracking technology and questionnaire-based survey on cruise passengers visiting the city of Palermo are used, to identify the main determinants which characterize their stop locations pattern. The spatial intensity of stop locations is estimated through a Gibbs point process model, taking into account for both individual-related variables, contextual-level information, and for spatial interaction among stop points. The Berman-Turner device for maximum pseudolikelihood is considered, by using a quadrature scheme generated on the network. The approach used allows taking into account the linear network determined by the street configuration of the destination under analysis. The results show an influence of both socio-demographic and trip-related characteristics on the stop location patterns, as well as the relevance of distance from the main attractions, and potential interactions among cruise passengers in stop configuration. The proposed approach represents both improvements from the methodological perspective, related to the modelling of spatial point process on a linear network, and from the applied perspective, given that better knowledge of the determinants of spatial intensity of visitors’ stop locations in urban contexts may orient destination management policy.

scholarly journals PI-Bully: Personalized Cyberbullying Detection with Peer Influence

2019 ◽  
Author(s):  
Lu Cheng ◽  
Jundong Li ◽  
Yasin Silva ◽  
Deborah Hall ◽  
Huan Liu
Keyword(s):  
Peer Influence ◽  
Real World ◽  
Classification Model ◽  
Learning Models ◽  
Individual User ◽  
Collaborative Environment ◽  
Accurate Detection ◽  
Real World Datasets ◽  
Cyberbullying Detection

Cyberbullying has become one of the most pressing online risks for adolescents and has raised serious concerns in society. Recent years have witnessed a surge in research aimed at developing principled learning models to detect cyberbullying behaviors. These efforts have primarily focused on building a single generic classification model to differentiate bullying content from normal (non-bullying) content among all users. These models treat users equally and overlook idiosyncratic information about users that might facilitate the accurate detection of cyberbullying. In this paper, we propose a personalized cyberbullying detection framework, PI-Bully, that draws on empirical findings from psychology highlighting unique characteristics of victims and bullies and peer influence from like-minded users as predictors of cyberbullying behaviors. Our framework is novel in its ability to model peer influence in a collaborative environment and tailor cyberbullying prediction for each individual user. Extensive experimental evaluations on real-world datasets corroborate the effectiveness of the proposed framework.

Sign in / Sign up

Export Citation Format

Share Document