16 Questions 157 Answers 0 Followers
Questions related from Sergey Porotsky
Some feature of my classification task can get only limited amount of values, e.g. only values 0 and 1. What is your opinion and experience: is it useful to fix values of this feature and to solve...
05 May 2019 4,812 7 View
On the traditional statistics tasks simultaneously with Probability Value calculation usually we can also estimate Confidence Interval for this Probability value (with some pre-defined Confidence...
04 April 2019 9,304 3 View
Support Vector Machines (SVM), Logistic Regression (LR) and some other classification methods can guarantee, that obtained solution really will be global minimum, because they use convex goal...
11 November 2018 8,135 6 View
Direct using of the Matlab Neural Network Toolbox for unbalanced Data Set (only 15% of all objects are positive) gets non-correct results : Sensitivity = 0.4%, Specificity = 99%. Direct using of...
10 October 2018 3,557 7 View
To perform tuning (i.e. to select meta-parameter values) on Classification Tasks usually use such integral criteria, as F_k_measure ( = (k+1)/( k/recall +1/precision ) ), G_measure ( = (...
07 July 2018 7,837 3 View
I've read several times that on the problems of large dimension (Image Recognition, Text Mining, ...), the Deep Learning method gives significantly higher accuracy than the "classical" methods...
05 May 2018 3,359 15 View
In my opinion, well-known classification methods (NN, SVM, LogRegr., etc.) work good only for tasks with continuous variables or discrete variables with a lot of possible values (e.g., “age”). If...
05 May 2018 7,067 5 View
Some Machine Learning methods have versions both for Classification and Regression. For example, SVM method (Support Vector Machine) has realizations SVC (Support Vector Classification) and SVR...
02 February 2018 6,474 14 View
Assume, that x is input parameter, y is real output value and z is predicted output value. Early (20 years ago!) for sensitivity I have mean indicator of correction of prediction function changing...
12 December 2017 5,851 5 View
To check, how our statistics corresponds for selected distribution, we should perform Goodness-of-Fit test. Usually these tests are Chi-Square, Kolmogorov-Smirnov, Kramer-Mizes and etc. But is it...
11 November 2017 518 24 View
Values of some parameters for some objects are missing (blank). How to use (transform) SVM classification method for such Data Sets? Please, give me advises, links, references, articles, etc....
05 May 2017 4,645 14 View
Dear colleges, Hello! We are going to extend our MLE (Maximum Likelihood Estimation)- based tools of parameters calculation for GRP - early we solved this task only for “classic” repairable...
08 August 2016 9,123 4 View
Hello! For Global Optimization I often use random-search oriented methods, such as Genetic Alg., Cross-Entropy Alg., etc. I always see, that accuracy and speed of convergence very large depend of...
07 July 2016 8,335 11 View
Hello! I am only starting to study GRP (General Renewal Process), so sorry, if my question will be too simple -:). Main expression for GRP is A[i] - A[i-1] = q*(S[i] - S[i-1]), where A is Virtual...
07 July 2016 4,250 4 View
Hello! I understand and fully agree, that in many fields the Machine Learning methods get us better results for comparison with Traditional Statistical methods. But what is advantage to use...
05 May 2016 9,655 10 View
I should solve binary classification task. For training (and further for new object recognition) I can use two cohorts. First have M1 records (for N1 features + label), second has M2 records (for...
01 January 1970 4,356 12 View
