Topic modeling and improvement of image representation for large-scale image retrieval

Tu, Nguyen Anh; Dinh, Dong-Luong; Rasel, Mostofa Kamal; Lee, Young-Koo

doi:10.1016/j.ins.2016.05.029

Cited by 15 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other problems of NMF models were described in [ 4 , 5 ]. At the same time, despite broad usage of probabilistic topic models in different fields of machine learning [ 6 , 7 , 8 , 9 ], they, too, possess a set of problems limiting their usage for big data analysis.…”

Section: Introductionmentioning

confidence: 99%

Estimating Topic Modeling Performance with Sharma–Mittal Entropy

Koltcov

Ignatenko

Koltsova

2019

Entropy

View full text Add to dashboard Cite

Topic modeling is a popular approach for clustering text documents. However, current tools have a number of unsolved problems such as instability and a lack of criteria for selecting the values of model parameters. In this work, we propose a method to solve partially the problems of optimizing model parameters, simultaneously accounting for semantic stability. Our method is inspired by the concepts from statistical physics and is based on Sharma–Mittal entropy. We test our approach on two models: probabilistic Latent Semantic Analysis (pLSA) and Latent Dirichlet Allocation (LDA) with Gibbs sampling, and on two datasets in different languages. We compare our approach against a number of standard metrics, each of which is able to account for just one of the parameters of our interest. We demonstrate that Sharma–Mittal entropy is a convenient tool for selecting both the number of topics and the values of hyper-parameters, simultaneously controlling for semantic stability, which none of the existing metrics can do. Furthermore, we show that concepts from statistical physics can be used to contribute to theory construction for machine learning, a rapidly-developing sphere that currently lacks a consistent theoretical ground.

show abstract

Section: Introductionmentioning

confidence: 99%

Estimating Topic Modeling Performance with Sharma–Mittal Entropy

Koltcov

Ignatenko

Koltsova

2019

Entropy

View full text Add to dashboard Cite

show abstract

“…Representation learning refines the input raw data by highlighting useful informa-135 tion and eliminating redundant information and noise. It is one of the most important techniques in computer vision and multimedia [5,7], and so far deep learning is the most successful representation learning technique [16,3,34]. One of the most commonly used deep representation learning methods is the convolutional neural network (CNN) [17], which is widely used [29,13,44,18].…”

Section: Representation Learningmentioning

confidence: 99%

Attention driven multi-modal similarity learning

Gao

Goulermas

et al. 2018

Information Sciences

View full text Add to dashboard Cite

To learn a function for measuring similarity or relevance between objects is an important machine learning task, referred to as similarity learning. Conventional methods are usually insufficient for processing complex patterns, while more sophisticated methods produce results supported by parameters and mathematical operations that are hard to interpret. To improve both model robustness and interpretability, we propose a novel attention driven multi-modal algorithm, which learns a distributed similarity score over different relation modalities and develops an interaction-oriented dynamic attention mechanism to selectively focus on salient patches of objects of interest. Neural networks are used to generate a set of high-level representation vectors for both the entire object and its segmented patches. Multi-view local neighboring structures between objects are encoded in the high-level object representation through an unsupervised pre-training procedure. By initializing the relation embeddings with object cluster centers, each relation modality can be reasonably interpreted as a semantic topic. A layer-wise training scheme based on a mixture of unsupervised and supervised training is proposed to improve generalization. The effectiveness of the proposed method and its superior performance compared against state-of-the-art algorithms are demonstrated through evaluations based on different image retrieval tasks.

show abstract

“…Topic modeling (TM) is a machine learning algorithm that allows for automatic extraction of topics from large text data. Nowadays, TM is widely used in different research fields such as social sciences [ 1 ], historical science [ 2 ], linguistics [ 3 ], literary studies [ 4 ], mass spectrometry [ 5 ], and image retrieval, among others [ 6 ]. However, to model a dataset, most of the topic models require the TM user to select the number of topics that, in practice, is an ambiguous and complex task.…”

Section: Introductionmentioning

confidence: 99%

Renormalization Analysis of Topic Models

Koltcov

Ignatenko

2020

Entropy

View full text Add to dashboard Cite

In practice, to build a machine learning model of big data, one needs to tune model parameters. The process of parameter tuning involves extremely time-consuming and computationally expensive grid search. However, the theory of statistical physics provides techniques allowing us to optimize this process. The paper shows that a function of the output of topic modeling demonstrates self-similar behavior under variation of the number of clusters. Such behavior allows using a renormalization technique. A combination of renormalization procedure with the Renyi entropy approach allows for quick searching of the optimal number of topics. In this paper, the renormalization procedure is developed for the probabilistic Latent Semantic Analysis (pLSA), and the Latent Dirichlet Allocation model with variational Expectation–Maximization algorithm (VLDA) and the Latent Dirichlet Allocation model with granulated Gibbs sampling procedure (GLDA). The experiments were conducted on two test datasets with a known number of topics in two different languages and on one unlabeled test dataset with an unknown number of topics. The paper shows that the renormalization procedure allows for finding an approximation of the optimal number of topics at least 30 times faster than the grid search without significant loss of quality.

show abstract

Topic modeling and improvement of image representation for large-scale image retrieval

Cited by 15 publications

References 21 publications

Estimating Topic Modeling Performance with Sharma–Mittal Entropy

Estimating Topic Modeling Performance with Sharma–Mittal Entropy

Attention driven multi-modal similarity learning

Renormalization Analysis of Topic Models

Contact Info

Product

Resources

About