Revealing Prevailing Semantic Contents of Clusters Generated from Untagged Freely Written Text Documents in Natural Languages

Žižka, Ján; Dařena, František

doi:10.1007/978-3-642-40585-3_55

Cited by 5 publications

(2 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mapping the document vectors to the values of stock prices might be related to many problems. The number of attributes describing texts is generally very high (can be in the order of tens of thousands) and the attribute vectors are sparse (Žižka and Dařena, 2013). A certain type of event can be also characterized by many different words.…”

Section: Methodsmentioning

confidence: 99%

“…The clusters are expected to be homogeneous inside and must be clearly distinguishable from each other to express their own distinct information. Clustering has been successfully applied for organizing and searching large text collections (Bsoul et al, 2013;Dhillon and Modha, 1999;Guo and Zhang, 2009;Tseng et al, 2007) and can be used to discover main topics hidden in collections of texts (Žižka and Dařena, 2013;Barák et al, 2015).…”

Section: Processing Documentsmentioning

confidence: 99%

See 1 more Smart Citation

Analysis of the Association between Topics in Online Documents and Stock Price Movements

Dařena

Přichystal

2018

Acta Univ. Agric. Silvic. Mendelianae Brun.

Self Cite

View full text Add to dashboard Cite

This paper aims at discovering the topics hidden in the newspaper articles that have an impact on movements of stock prices of the corresponding companies. Document topics are characterized by combinations of specific words in documents and are shared across a document collection. We describe the process of discovering the topics, the creation of a mapping of the topics to stock price movements, and quantifying and evaluating the results. As the method for finding and quantifying the association, we use machine learning‑based classification. We achieved an accuracy of stock price movement predictions higher than 70 %. A feature selection procedure was applied to the features characterizing the topics in order to facilitate the process of assigning a label to the topic by a human expert.

show abstract

Section: Methodsmentioning

confidence: 99%