News topics categorization using latent Dirichlet allocation and sparse representation classifier

Lee, Yuan-Shan; Lo, Rocky; Chen, Chia‐Yen; Lin, Po-Chuan; Wang, Jia-Ching

doi:10.1109/icce-tw.2015.7216819

Cited by 17 publications

(14 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It is proved that solving minimum reduction of decision table is a NP hard problem [10]. Therefore, heuristic algorithms are used for obtaining suboptimal solutions.…”

Section: Decision Table Reductionmentioning

confidence: 99%

Dimensionality Reduction by Feature Co-Occurrence based Rough Set

La¹

2019

IJPE

View full text Add to dashboard Cite

Feature selection is the key issue of unstructured data mining related fields. This paper presents a dimensionality reduction method which uses a rough set as the feature selection tool. Different from previous rough set based classification algorithm, it takes feature cooccurrence into account when make attribution reduction to get a more accurate feature subset. The novel method called Feature Cooccurrence Quick Reduction algorithm is in this article. Experimental results show it has a high efficiency in dimensionality reductiontime consumption by approximately 23% less than traditional rough set based dimensionality reduction methods. Moreover, classification based on the feature set selected by Feature Co-occurrence Quick Reduction algorithm is more precise. The proposed algorithm is helpful to us for refining knowledge from massive unstructured data.

show abstract

“…It is proved that solving minimum reduction of decision table is a NP hard problem [10]. Therefore, heuristic algorithms are used for obtaining suboptimal solutions.…”

Section: Decision Table Reductionmentioning

confidence: 99%

Dimensionality Reduction by Feature Co-Occurrence based Rough Set

La¹

2019

IJPE

View full text Add to dashboard Cite

show abstract

“…Em [Feuerriegel et al 2016], os autores tentaram identificar os efeitos de tópicos escondidos em notícias do setor financeiro, como preços de ações, usando LDA. Em uma linha semelhante, o trabalho de [Lee et al 2015] fez uma categorização de tópicos de notícias usando modelos de LDA e uma representação esparsa para aprimorar sistemas de recomendação de notícias. [Qian et al 2015] aplicou uma abordagem LDA multi-modal supervisionada para auxiliar na classificação de eventos de um grande número de dados gerados por usuários, dispostos em mídias sociais.…”

Section: Trabalhos Correlatosunclassified

Melhorando Sistemas de Social CRM por meio de Eletronic Word-of-Mouth

Netto

Almeida

Lobato

et al. 2019

REIC

View full text Add to dashboard Cite

A Web 2.0 modificou radicalmente a forma pela qual os consumidores tomam decisões de compra, necessitando de uma nova abordagem de gerenciamento de dados de redes sociais, conhecida por Social Customer Relationship Management. Além disso, nesses novos ambientes surgiu o boca-a-boca eletrônico, sendo a capacidade de disponibilizar opiniões sobre produtos e serviços por meio de redes sociais com qualquer outro usuário. Consequentemente, dentro dessas redes residem dados gerados por usuários em forma de comentários e reclamações que, sob análise, podem trazer em vantagens competitivas para empresas. Os resultados deste trabalho confirmam que a ferramenta é uma inovação tecnológica, que permiti uma melhor interação empresa-consumidor, gerando maior confiabilidade nos processos decisórios de vendas e de atendimento aos clientes.

show abstract

“…Because α produces topic distribution θ, which identifies the specific topic, and β produces word distribution ϕ, which identifies the specific word. Therefore, Equation (1) is equivalent to the joint probability distribution p of all the variables expressed by Equation (2).…”

Section: Topic Modelmentioning

confidence: 99%

“…How to effectively manage the surge of image data and obtain semantic information from these data has become an urgent task. In recent years, the model of latent Dirichlet allocation [1] has been successfully and rapidly applied to many fields by processing the texts or images to obtain thematic variables, and use them as a basis of classification or other processing, such as processing of texts [2], image retrieval [3], remote sensing images [4], data mining [5], and so on. However, the LDA model has higher requirements for image data, and the change of brightness, illumination, and scale of images will bring great difficulty to image recognition.…”

Section: Introductionmentioning

confidence: 99%

Scene Semantic Recognition Based on Probability Topic Model

Feng

2018

Information

View full text Add to dashboard Cite

In recent years, scene semantic recognition has become the most exciting and fastest growing research topic. Lots of scene semantic analysis methods thus have been proposed for better scene content interpretation. By using latent Dirichlet allocation (LDA) to deduce the effective topic features, the accuracy of image semantic recognition has been significantly improved. Besides, the method of extracting deep features by layer-by-layer iterative computation using convolutional neural networks (CNNs) has achieved great success in image recognition. The paper proposes a method called DF-LDA, which is a hybrid supervised-unsupervised method combined CNNs with LDA to extract image topics. This method uses CNNs to explore visual features that are more suitable for scene images, and group the features of salient semantics into visual topics through topic models. In contrast to the LDA as a tool for simply extracting image semantics, our approach achieves better performance on three datasets that contain various scene categories.

show abstract

News topics categorization using latent Dirichlet allocation and sparse representation classifier

Cited by 17 publications

References 2 publications

Dimensionality Reduction by Feature Co-Occurrence based Rough Set

Dimensionality Reduction by Feature Co-Occurrence based Rough Set

Melhorando Sistemas de Social CRM por meio de Eletronic Word-of-Mouth

Scene Semantic Recognition Based on Probability Topic Model

Contact Info

Product

Resources

About