M
Moshe Lichman
Researcher at University of California, Irvine
Publications - 7
Citations - 239
Moshe Lichman is an academic researcher from University of California, Irvine. The author has contributed to research in topics: Population & Mixture model. The author has an hindex of 4, co-authored 7 publications receiving 198 citations. Previous affiliations of Moshe Lichman include Xerox & University of California.
Papers
More filters
Proceedings ArticleDOI
Modeling human location data with mixtures of kernel densities
Moshe Lichman,Padhraic Smyth +1 more
TL;DR: The experimental results indicate that the mixture-KDE method provides a useful and accurate methodology for capturing and predicting individual-level spatial patterns in the presence of noisy and sparse data.
Journal ArticleDOI
Using Social Media to Measure Temporal Ambient Population: Does it Help Explain Local Crime Rates?
TL;DR: In this article, the spatial and temporal confluence of offenders and targets is considered for studies assessing routine activities theory, where people move about during the daytime and night for routine activities.
Journal ArticleDOI
Predicting Consumption Patterns with Repeated and Novel Events
TL;DR: This work finds that widely-used matrix factorization methods are limited in their ability to capture important details in historical behavior, resulting in relatively low predictive accuracy for these types of problems, and proposes an interpretable and scalable mixture model framework that balances individual preferences in terms of exploration and exploitation.
Proceedings ArticleDOI
Prediction of Sparse User-Item Consumption Rates with Zero-Inflated Poisson Regression
Moshe Lichman,Padhraic Smyth +1 more
TL;DR: This paper uses zero-inflated Poisson (ZIP) regression models as the basis for the modeling approach, leading to a general framework for modeling user-item consumption rates over time and shows that these models are more flexible in capturing user behavior than alternatives such as well-known latent factor models based on matrix factorization.
Proceedings ArticleDOI
Personalized location models with adaptive mixtures
TL;DR: A general mixture model framework for learning individual-level location models where the model adaptively combines different types of smoothing information and can be significantly more accurate than more traditional smoothing and matrix factorization techniques.