Katarzyna Kaczmarek, Ph.D.

Katarzyna Kaczmarek has developed the following methods and tools:

  1. Bayesian Regression with Linguistic Knowledge (BRLK), a method to predict sequences of linguistic information comprising data similarity analysis, time series segmantation, imprecise expressions processing and Bayesian inference
  2. Model Identification with Linguistic Summaries (MILS), a method to construct probability distributions based on machine learning algorithms, advanced segmentation techniques and linguistic summarization of time series
  3. Bayesian Forecasting with Soft Computing Prior Information (BFSC), a method of Bayesian prediction for short time series which employs imprecise expert knowledge on expected trends concerning forecasted phenomena
  4. Granular Computing framework for the Bayesian Forecasting (GCBF), a generalization of the BFSC method for long time series using imprecise expert knowledge, more effective machine learning algorithms such as SVM or k-NN, time series segmentation and linguistic summarization
  5. Classification with Linguistic Summaries (CLS), a method to classify a time series using the linguistic summarization results
  6. an implemenation of the above mentioned methods forming a decision support system applicable to time series analysis and prediction.

Ms. Katarzyna Kaczmarek published the results of her work in many papers (cf., here) two of which has been distinguished at two conferences (FedCSIS 2013 and ACHI 2015) with a Best Paper Award.

Ms. Katarzyna Kaczmarek defended her PhD thesis entitled "Soft Computing Methods in Bayesian Analysis of Time Series" and has been granted the PhD degree with distinction in computer science  on March 5, 2015.

You are here: Home Results Katarzyna Kaczmarek, Ph.D.