Confidence Measure for Czech Document Classification

Král, Pavel; Lenc, Ladislav

doi:10.1007/978-3-319-18117-2_39

Cited by 3 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi-Label Classification The Czech text document corpus CTDC (Kral and Lenc 2018) is a dataset designed for direct comparison of document classification on Czech data. It comprises 12,000 news articles labeled with 37 categories used for classification.…”

Section: Extrinsic Evaluationmentioning

confidence: 99%

Some Like It Small: Czech Semantic Embedding Models for Industry Applications

Bednář,

Náplava,

Barančíková

et al. 2024

AAAI

View full text Add to dashboard Cite

This article focuses on the development and evaluation of Small-sized Czech sentence embedding models. Small models are important components for real-time industry applications in resource-constrained environments. Given the limited availability of labeled Czech data, alternative approaches, including pre-training, knowledge distillation, and unsupervised contrastive fine-tuning, are investigated. Comprehensive intrinsic and extrinsic analyses are conducted, showcasing the competitive performance of our models compared to significantly larger counterparts, with approximately 8 times smaller size and 5 times faster speed than conventional Base-sized models. To promote cooperation and reproducibility, both the models and the evaluation pipeline are made publicly accessible. Ultimately, this article presents practical applications of the developed sentence embedding models in Seznam.cz, the Czech search engine. These models have effectively replaced previous counterparts, enhancing the overall search experience for instance, in organic search, featured snippets, and image search. This transition has yielded improved performance.

show abstract

Section: Extrinsic Evaluationmentioning

confidence: 99%

Some Like It Small: Czech Semantic Embedding Models for Industry Applications

Bednář,

Náplava,

Barančíková

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Mathematical regression model is a good assessment of the weight of the text feature [4]. It is starting from a set of sample data, to determine the mathematical relationship in the variables, and the credibility of those relationships is done by various statistical tests to find out the impact of a particular variable from the many variables affecting significant and no significant variables [5].…”

Section: Document Classification On Mathematical Regression Modelmentioning

confidence: 99%

Research on Document Content Classification on Mathematical Regression Model

Long¹,

Li²

2015

Proceedings of the 3rd International Conference on Mechatronics and Industrial Informatics

View full text Add to dashboard Cite

Abstract. To improve the document classification problem, this study proposes a classification algorithm based on mathematical regression model, making Chinese document classification get rid of the dependence on traditional dictionary method. The method of extracting high frequency keywords, establishes the appropriate matrix model, making a high-dimensional document change into a low-dimensional document, and then use mathematical regression model to give a comprehensive feature weighting function by corpus training. It explored an approach to avoid the traditional method of the problem of curse of dimensionality.

show abstract

“…As already stated, the confidence measure is used as a post-processing of the classification to identify whether the result is correct or not. In this work, we combine two measures based on the posterior class probability by a multi-layer perceptron classifier as already presented in [15].…”

Section: Confidence Measuresmentioning

confidence: 99%

SAPKOS: Experimental Czech Multi-label Document Classification and Analysis System

Lenc

Král

2015

IFIP Advances in Information and Communication Technology

Self Cite

View full text Add to dashboard Cite

This paper presents an experimental multi-label document classification and analysis system called SAPKOS. The system which integrates the state-of-the-art machine learning and natural language processing approaches is intended to be used by the Czech news Agency (ČTK). Its main purpose is to save human resources in the task of annotation of newspaper articles with topics. Another important functionality is automatic comparison of theČTK production with popular Czech media. The results of this analysis will be used to adapt theČTK production to better correspond to the today's market requirements. An interesting contribution is that, to the best of our knowledge, no other automatic Czech document classification system exists. It is also worth mentioning that the system accuracy is very high. This score is obtained due to the unique system architecture which integrates a maximum entropy based classification engine with the novel confidence measure method.

show abstract

Confidence Measure for Czech Document Classification

Cited by 3 publications

References 26 publications

Some Like It Small: Czech Semantic Embedding Models for Industry Applications

Some Like It Small: Czech Semantic Embedding Models for Industry Applications

Research on Document Content Classification on Mathematical Regression Model

SAPKOS: Experimental Czech Multi-label Document Classification and Analysis System

Contact Info

Product

Resources

About