Toward content based retrieval from scientific text corpora

Kloptchenko, Antonina; Back, Barbro; Visa, Ari; Toivonen, Jarmo; Vanharanta, Hannu

doi:10.1109/icais.2002.1048170

Cited by 3 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a result, the selection of prototypical texts has become an active area of research in natural language processing (NLP). Visa et al (2001) and Kloptchenko et al (2002Kloptchenko et al ( , 2004 adopt an approach that initially requires human judgment to pick prototypical texts from a larger corpus that represent different text types (or classes). Next, they analyse the prototype texts in terms of word and sentence structure, i.e.…”

Section: Text Prototype Selection Methodsmentioning

confidence: 99%

ProtAnt

Anthony

Baker

2015

IJCL

View full text Add to dashboard Cite

Corpus-based researchers and traditional qualitative researchers, such as those interested in critical discourse analysis, are often required to select prototypical texts for close reading that include the language features of interest that are present in a much larger corpus. Traditional approaches to this selection procedure have been largely ad hoc. In this paper, we offer a more principled way of selecting texts for close reading based on a ranking of texts in terms of the number of keywords they contain. To facilitate this analysis, we have developed a multiplatform, freeware software tool calledProtAntthat analyses the texts, generates a ranked list of keywords based on statistical significance and effect size, and then orders the texts by the number of keywords in them. We describe various experiments that demonstrate theProtAntanalysis is effective not only at identifying prototypical texts, but also identifying outlier texts that may need to be removed from a target corpus.

show abstract

Section: Text Prototype Selection Methodsmentioning

confidence: 99%

ProtAnt

Anthony

Baker

2015

IJCL

View full text Add to dashboard Cite

show abstract

“…Surprisingly, little research has been done on utilizing the textual content of annual reports to build predictive models, despite findings that the reports have the potential to serve as indicators of company future prospects. The most related work in this direction is that of Kloptchenko et al (2002) and Visa et al (2000). In Kloptchenko et al (2002), company quarterly reports and corresponding financial ratios were clustered with prototype matching clustering and SOM clustering respectively.…”

Section: Introductionmentioning

confidence: 99%

“…The most related work in this direction is that of Kloptchenko et al (2002) and Visa et al (2000). In Kloptchenko et al (2002), company quarterly reports and corresponding financial ratios were clustered with prototype matching clustering and SOM clustering respectively. Although the two clusters did not coincide, the authors found that changes in textual reports tended to occur ahead of the changes in financial performance.…”

mentioning

confidence: 99%

Exploring the Forecasting Potential of Company Annual Reports

Qiu

Srinivasan

Street

2006

Proc of Assoc for Info

View full text Add to dashboard Cite

Previous research indicates that the narration disclosure in company annual reportscan be used to assist in assessing the company's short-term financial prospects. However, not much effort has been made to systematically and automatically assess the predictive potential of such reports using text classification, information retrieval, and machine learning techniques. In this study, we built SVM-based predictive models with different feature selection methods from ten years of annual reports of 30 companies. We used feature selection methods to reduce the term space and studied the class-related vocabulary. Evaluation of predictive accuracy is performed with cross validation and t-test significance tests. We compare different models' performance and analyze misclassification rates by year and by industry. We identify the strengths and weaknesses of each model. Our results support the feasibility of automatically predicting next-year company financial performance from the current year's report. We suggest text features can be further studied to understand their roles as indicators of company's future performance. This research paves the way for large-scale automatic analysis of the relationship between annual reports and short-term performance, as well as the identification of interesting signals within annual reports. IntroductionCompany annual reports (10K filings) are freely available to the public and contain required disclosures, quantitative summaries of the company's financial performance as well as textual discussions. These reports are of great importance in helping investors, corporate managers, and financial analysts with their decision-making. Studies have shown that the narration sections of 10K filings provide information that is as useful as the financial ratios to financial analysts while predicting the company's future prospects (Roger & Grant,1997, Schipper, 1991. The SEC (Securities and Exchange Commission) also requires the reporting of the firm's strategies and managerial priorities, and its view of the past year's performance and future prospects. The major mandatory disclosures in annual reports include reasons for price and sales changes, reasons for revenue and cost changes, planned expenditures, known trends, and future liquidity positions.Annual reports have been studied as a marketing and communication tool that the corporation uses to convey an image or messages to its stakeholders (Herreman & Ryans, 1995). More recent studies on the relationship between the reports and firm performance have focused on special sections of the reports, such as the chairman's statement (Smith & Taffler, 2000), management discussion and analysis (MD&A) (Bryan, 1997), president's letter (Abrahamson & Amir, 1996) as well as the general writing style and readability (Subramanian et al., 1993). The methods these studies employ are generally semi-automatic, including content analysis, readability measurements, manual annotation and categorization, linear discriminant analysis, logit model and other statistical...

show abstract

An approach to online text generation for e-learning purposes

Mahmoudi

Badie

2004

International Conference on Information Technology: Coding and Computing, 2004. Proceedings. ITCC 2004.

View full text Add to dashboard Cite

Toward content based retrieval from scientific text corpora

Cited by 3 publications

References 11 publications

ProtAnt

ProtAnt

Exploring the Forecasting Potential of Company Annual Reports

An approach to online text generation for e-learning purposes

Contact Info

Product

Resources

About