Filters








459 Hits in 6.6 sec

Sequential Activity Profiling: Latent Dirichlet Allocation of Markov Chains

Mark Girolami, Ata Kabán
2005 Data mining and knowledge discovery  
The algorithm is based on a straightforward generalization of latent Dirichlet allocation to time-invariant Markov chains of arbitrary order.  ...  To provide a parsimonious generative representation of the sequential activity of a number of individuals within a population there is a necessary tradeoff between the definition of individual specific  ...  The consistent generative semantics of the recently introduced latent Dirichlet allocation (LDA) [3] will be adopted and by analogy with [16] the resulting model will be referred to as a simplicial  ... 
doi:10.1007/s10618-005-0362-2 fatcat:cd7bcqnh55fy3pnhqu4axzhpnq

RaptGen: A variational autoencoder with profile hidden Markov model for generative aptamer discovery [article]

Natsuki Iwano, Tatsuo Adachi, Kazuteru Aoki, Yoshikazu Nakamura, Michiaki Hamada
2021 bioRxiv   pre-print
RaptGen exploits a profile hidden Markov model decoder to represent motif sequences effectively.  ...  We demonstrated that RaptGen could be applied to activity-guided aptamer generation according to Bayesian optimization.  ...  p(x) can be written by using the Markov chain rule: 86 p (x) = π p (x, π) = p(x 0:L+1 , π last = M m+1 ), 3050 - 3050 AV), respectively.  ... 
doi:10.1101/2021.02.17.431338 fatcat:hukpewsdc5bn5nob622myx76ve

Exploring the Sensory Profiles of Children on the Autism Spectrum Using the Short Sensory Profile-2 (SSP-2)

Kate Simpson, Dawn Adams, Clair Alston-Knox, Helen S. Heussler, Deb Keen
2019 Journal of Autism and Developmental Disorders  
The aim of this study was to identify sensory subtypes in children on the autism spectrum using the Short Sensory Profile-2 (SSP-2).  ...  Analysis using Dirichlet process mixture model identified a two-cluster model which provided the best solution to subtype sensory responses.  ...  Essentially, the DPMM has an infinite number of components, and the algorithm implemented with Monte Carlo Markov Chain (MCMC) techniques adapts the number of "active" components based on the previous  ... 
doi:10.1007/s10803-019-03889-2 fatcat:lh5uzcmhm5bjrjk7g37zmmnvxy

Sequential latent Dirichlet allocation

Lan Du, Wray Buntine, Huidong Jin, Changyou Chen
2011 Knowledge and Information Systems  
In this article, we address this problem by presenting a novel variant of latent Dirichlet allocation (LDA): Sequential LDA (SeqLDA).  ...  Such progressive sequential dependency is captured by using the hierarchical two-parameter Poisson-Dirichlet process (HPDP).  ...  NICTA is funded by the Australian Government as represented by the Department of Broadband, Communications and the Digital Economy and the Australian Research Council through the ICT Centre of Excellence  ... 
doi:10.1007/s10115-011-0425-1 fatcat:avtzeaozsvb25eudm372mdfjim

Human Activity Clustering for Online Anomaly Detection

Xudong Zhu, Zhijing Liu, Juehui Zhang
2011 Journal of Computers  
This paper aims to address the problem of profiling human activities captured in surveillance videos for the applications of online normal human activity recognition and anomaly detection.  ...  The framework consists of the following key components: 1) A compact and effective activity representation method is developed based on a stochastic sequence of spatiotemporal actions. 2) The natural grouping  ...  on discovering the natural grouping of activity using Hidden Markov Model with Latent Dirichlet Allocation (HMM-LDA).  ... 
doi:10.4304/jcp.6.6.1071-1079 fatcat:m7jtw33qezdchfx56z5xv2teay

A Survey on Journey of Topic Modeling Techniques from SVD to Deep Learning

Deepak Sharma, Bijendra Kumar, Satish Chand
2017 International Journal of Modern Education and Computer Science  
Here we present a survey on journey of topic modeling techniques comprising Latent Dirichlet Allocation (LDA) and non-LDA based techniques and the reason for classify the techniques into LDA and non-LDA  ...  These techniques reveal the hidden thematic structure in a collection of documents and facilitate to build up new ways to browse, search and summarize large archive of texts.  ...  LDA based sequence of words topic models with unsupervised learning a. Sequential Latent Dirichlet Allocation (seqLDA) Fig. 13 . 13 seqLDA Graphical Model D.  ... 
doi:10.5815/ijmecs.2017.07.06 fatcat:nadnmsoj4zdi7onlxivrne6gqm

Combining eye movements and collaborative filtering for proactive information retrieval

Kai Puolamäki, Jarkko Salojärvi, Eerika Savia, Jaana Simola, Samuel Kaski
2005 Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '05  
Collaborative filtering is carried out using the User Rating Profile model, a state-of-the-art probabilistic latent variable model, computed using Markov Chain Monte Carlo techniques.  ...  The best prediction accuracy still leaves room for improvement but shows that proactive information retrieval and combination of many sources of relevance feedback is feasible.  ...  This work was supported by the Academy of Finland, decision no. 79017, and by the IST Programme of the European Community, under the PASCAL Network of Excellence, IST-2002-506778.  ... 
doi:10.1145/1076034.1076062 dblp:conf/sigir/PuolamakiSSSK05 fatcat:cqfmykb3b5bxrjr2gigpt3uek4

WWW 2011 invited tutorial overview

Amr Ahmed, Alexander Smola
2011 Proceedings of the 20th international conference companion on World wide web - WWW '11  
Secondly, the problems arising on the internet do not always fit well into the known categories for latent variable inference such as Latent Dirichlet Allocation or clustering.  ...  Firstly, we present a variety of applications ranging from general purpose document analysis, ideology detection, clustering of sequential data, and dynamic user profiling to recommender systems and data  ...  Latent Dirichlet Allocation (LDA) is a suitable tool for uncovering such dependencies. It proved successful on smaller scale datasets for extraction of topics in documents.  ... 
doi:10.1145/1963192.1963311 dblp:conf/www/AhmedS11 fatcat:t6ua5wckjjgtpaeubhhsh2ivra

Topic Modeling for Sequences of Temporal Activities

Zhiyong Shen, Ping Luo, Yuhong Xiong, Jun Sun, Yidong Shen
2009 2009 Ninth IEEE International Conference on Data Mining  
This paper presents an LDA-style topic model for sequences of temporal activities that captures three features of such sequences: 1) the counts of unique activities, 2) the Markov transition dependence  ...  This paper presents an LDAstyle topic model for sequences of temporal activities that captures three features of such sequences: 1) the counts of unique activities, 2) the Markov transition dependence  ...  [7] successfully employ Latent Dirichlet Allocation (LDA) [3] to discovery patterns in the sequences of daily human activities, in which LDA only models the counts of unique words (Feature 1) in a  ... 
doi:10.1109/icdm.2009.83 dblp:conf/icdm/ShenLXSS09 fatcat:yqmrmlwvprailnroil4xulykcq

Classification and Clustering Methods for Multiple Environmental Factors in Gene–Environment Interaction

Yi-An Ko, Bhramar Mukherjee, Jennifer A. Smith, Sharon L. R. Kardia, Matthew Allison, Ana V. Diez Roux
2016 Epidemiology  
., k-means, latent class analysis, classification and regression trees, Bayesian clustering using Dirichlet Process) to define subgroups corresponding to distinct environmental exposure profiles.  ...  These tools not only allow researchers to consider several environmental exposures in G×E analysis but also provide some insight into how genes modify the effect of a comprehensive exposure profile instead  ...  The funder was not involved in the preparation of the manuscript or in the final decision to publish.  ... 
doi:10.1097/ede.0000000000000548 pmid:27479650 pmcid:PMC5039086 fatcat:k4nynx24wrfujcvbr64tishx7i

Non-Employment Activity Type Imputation from Points of Interest and Mobility Data at an Individual Level: How Accurate Can We Get?

Thanos Bantis, James Haworth
2019 ISPRS International Journal of Geo-Information  
This research proposes a methodological framework for urban activity type inference using a Dirichlet multinomial dynamic Bayesian network with an empirical Bayes prior that can be applied to mobility  ...  The results provide evidence of the limits of activity detection accuracy using such data as determined by the Area Under Receiving Operating Curve (AUROC), log-loss, and accuracy metrics.  ...  The accuracy of the proposed method was also benchmarked against two popular activity type models, Hidden Markov Model (HMM) and Latent Dirichlet Allocation (LDA).  ... 
doi:10.3390/ijgi8120560 fatcat:nuasif3mpzb47nekics267yn5u

Toward Personalized Context Recognition for Mobile Users

Baoxing Huai, Enhong Chen, Hengshu Zhu, Hui Xiong, Tengfei Bao, Qi Liu, Jilei Tian
2014 ACM Transactions on Knowledge Discovery from Data  
Specifically, we first exploit the Bayesian Hidden Markov Model (B-HMM) for modeling context in the form of probabilistic distributions and transitions of raw context data.  ...  Also, we propose a sequential model by extending B-HMM with the prior knowledge of contextual features to model context more accurately.  ...  The proposed approach is based on the Latent Dirichlet Allocation topic model and is scalable for multiple contextual features.  ... 
doi:10.1145/2629504 fatcat:x3nrhg24mjdmrktosub3dx6axi

Bayesian non-parametric hidden Markov models with applications in genomics

C. Yau, O. Papaspiliopoulos, G. O. Roberts, C. Holmes
2010 Journal of The Royal Statistical Society Series B-statistical Methodology  
The algorithm involves analytic marginalizations of latent variables to improve the mixing, facilitated by exchangeability properties of the Dirichlet process that we uncover in the paper.  ...  analysis of parametric hidden Markov models.  ...  It is interesting to investigate whether our approach of integrating out the global allocation variables when updating the hidden Markov chain, which is based on proposition 1, can be extended in this  ... 
doi:10.1111/j.1467-9868.2010.00756.x pmid:21687778 pmcid:PMC3116623 fatcat:5xdtn4m2ifdbpn35wt3ey7dhlm

Spatiotemporal Sequential Influence Modeling for Location Recommendations

Jia-Dong Zhang, Chi-Yin Chow
2015 ACM Transactions on Intelligent Systems and Technology  
a user visiting a new location through the developed additive Markov chain that considers the effect of all visited locations in the check-in history of the user on the new location.  ...  However, human movement also exhibits spatiotemporal sequential patterns, but only few current studies consider spatiotemporal sequential influence of locations on users' check-in behaviors.  ...  This method ] exploits the well-known topic model, i.e., latent Dirichlet allocation, to infer personal interest and local preference (i.e., local specialty).  ... 
doi:10.1145/2786761 fatcat:ssogeviapjhtfirs6yksmhagom

TraLFM: Latent Factor Modeling of Traffic Trajectory Data

Meng Chen, Xiaohui Yu, Yang Liu
2019 IEEE transactions on intelligent transportation systems (Print)  
Thus, TraLFM models the joint action of sequential, personal and temporal factors in a unified way, and brings a new perspective to many applications such as latent factor analysis and next location prediction  ...  The widespread use of positioning devices (e.g., GPS) has given rise to a vast body of human movement data, often in the form of trajectories.  ...  Both probabilistic latent semantic analysis(PLSA) [18] and latent dirichlet allocation (LDA) [7] have been popular methods for exploratory analysis of text.  ... 
doi:10.1109/tits.2019.2912075 fatcat:2hn7fq5bojgs7jb5qpxjulluja
« Previous Showing results 1 — 15 out of 459 results