Skip to content

These implementations make use of an algorithm called DPD (Discrete Probability Detector), that transforms any sequence of symbols into a transition matrix. DPD is able to detect the number of states from the sequence and calculate the transition probabilities between these states. This version of DPD is made in JavaScript, VBS and VB6.

License

Notifications You must be signed in to change notification settings

Gagniuc/Discrete-Probability-Detector

Repository files navigation

Discrete Probability Detector

These applications use an algorithm called DPD (Discrete Probability Detector) that transforms any sequence of symbols into a transition matrix. The algorithm may receive special characters from the entire ASCII range. These characters can be letters, numbers or special characters (ie. `7#Eu9f$*"). The number of symbol/character types that make up a string, represent the number of states in a Markov chain. Thus, DPD is able to detect the number of states from the sequence and calculate the transition probabilities between these states. The final result of the algorithm is represented by a transition matrix (square matrix) which contains the transition probabilities between these symbol types (or states). The transition matrix can be further used for different prediction methods, such as Markov chains or Hidden Markov Models. This version of DPD is made in HTML/JavaScript/CSS, VB6, VBA (Excel).

Live demo

https://gagniuc.github.io/Discrete-Probability-Detector/

Screenshot

References

  • Paul A. Gagniuc. Markov chains: from theory to implementation and experimentation. Hoboken, NJ, John Wiley & Sons, USA, 2017, ISBN: 978-1-119-38755-8.

About

These implementations make use of an algorithm called DPD (Discrete Probability Detector), that transforms any sequence of symbols into a transition matrix. DPD is able to detect the number of states from the sequence and calculate the transition probabilities between these states. This version of DPD is made in JavaScript, VBS and VB6.

Topics

Resources

License

Stars

Watchers

Forks

Sponsor this project

 

Packages

No packages published

Languages