The problem of approximate string matching is important in many different areas such as computational biology, text processing and pattern recognition. I want to compare a newly developed string matching algorithm with standart algorithms such as Jaro, Jaro-Winkler, Levenshtein, Monge-Elkan, Smith Waterman etc.
I need experimental datasets for algoritm evaluation purposes...
Have you considered using an artificial grammar with noise to provide a useful dataset? I've used Reber's artificial grammar in cognitive models of intuitive (subconscious) learning. (Reber (1967) Implicit learning of artificial grammars. Journal of Verbal learning and Verbal Behavior, 6, pp 855-893.
Thanks Osman. But I think, there is a misunderstanding. I need datasets for experimental to compare my algorithm.
See Antoon Bronselaer and Guy De Tré, Properties of possibilistic string comparison (2010) IEEE TRANSACTIONS ON FUZZY SYSTEMS. 18(2). p.312-325). You will find our technique there and also the datasets we have used.
You may find spelling correction or short text normalisation (Twitter, SMS) corpora to evaluate your algorithm. I expect that some of them have gold standard normalised versions.
Dear Prof. Kilinc,
I have a suggestion which is to use any panel dataset of financial markets with corporate variables so that you can Dirichlet Differentiate the panel in amanner tested in my coauthored paper on mathematical statistical pricing published in Nature and listed on my RG page. Generating the Lie Algebra requires some reduction methods using Arrow DEbreu theory and Gauss Markov Procedures with set inclusion and exclusion principles for differentiation and integration to ensure D-Branes String closure properties of the estimation algorithm. It uses multidimensional approxiamation which are driven to convergence by zero probability of errors in probability limits using tensor products of directed graph groups. It may be of help in your case. Earl Chair Prof. Dr. SKM QC EPS Fellow (In) MES MRES MAICTE
Dear Prof. Killinc,
We have also used another experimental dataset on Drugs & Pharmaceuticals Pricing in Indian Market Group Homology after globalisation and experimental increase in number and variety of Drugs and Drug companies and found them to be satisfying the conditions necessary for Genetic String Theoretic Mallick-Hamburger- Mallick Dbranes Functor Algebra Calculus (also Econophysical Calculus) (Mallick (1998, Mallick, Hamburger, Mallick (2017, 2017)) community disease vector spacextime) drug delivery programmes algorithms even internationally (Mallick, Hamburger, Mallick (2017)). This is also related to Drugs & Pharmaceuticals Habit persistence amongst localised group set inclusion and exclusion, studied extensively by the Swedish School of Social Medicine (Social Sciences in general). It is shown to be consistent with patient and treatment cybernetic networks using the internet and also for Insurance Medicine ( Mallick, Hmaburger, Mallick (2017)).We have also shown elsewhere starting with Dutta & Mallick (1988) which has been archived by the Bnak-Fund Library, that this Public Health Delivery Physics and Engineering programme is in conformity with Biostatistical Probability calculations of life-expectancy and life-value with applications to disease control programmes and social security.
Soumitra K. Mallick
for Soumitra K. Mallick, Nick Hamburger, Sandipan Mallick.
- Badges
- Science topic
- Similar topics
- Books