Transactions on Machine Learning and Artificial Intelligence

Electrical Conductance Analysis of Solanum Lycopersicum under Biotic Stress

2020-05-13T18:46:33+01:00

Our purpose is to provide different parameters of control from which one can identify a sick plant before the appearance of the first symptoms. We made a stochastic analysis and an analysis according to the theory of information, to deduce those characteristics parameters. It came out from our analysis that the DSP of health plant is above the DSP of the sick plant. Generally, the DSP of health and treated plant is above the DSP of sick and treated plant. However there is an overlapping between the DSP of sick and treated plant, and the health one for the whole value of the normalized reduced frequency. The average conductance of health plant is higher than the average conductance of sick plant. We also observed that, average conductance of health and treated plant is lower than the average conductance of sick and treated plant. The standard deviation of health plant is higher than the standard deviation of sick plant. We also observed that, standard deviation of health and treated plant is lower than the standard deviation of sick and treated plant. The electric conductance signal G(ω,t) of Solanum lycopersicum leaf plant is not a statistics process in the broad sense (SSL). Electric conductance G(ω,t) of the plant is a non ergotic signal. The entropy of the sick plant is higher than the entropy of the health one. Those parameters can be used during the development of informatics application, and can be used in I.O.T. (internet of thing)

Primality test and primes enumeration using odd numbers indexation

2020-05-13T18:46:27+01:00

Odd numbers can be indexed by the map k(n)=(n-3)⁄2,n∈2N+3. We first propose a basic primality test using this index function that was first introduced in [8]. Input size of operations is reduced which improves computational time by a constant. We then apply similar techniques to Atkin’s prime-numbers sieve which uses modulus operations and finally to Pritchard’s wheel sieve, in both case yielding similar results.

COSM: Controlled Over-Sampling Method

2020-05-13T18:46:30+01:00

The class imbalance problem is widespread in Data Mining and it can reduce the general performance of a classification model. Many techniques have been proposed in order to overcome it, thanks to which a model able to handling rare events can be trained. The methodology presented in this paper, called Controlled Over-Sampling Method (COSM), includes a controller model able to reject new synthetic elements for which there is no certainty of belonging to the minority class. It combines the common Machine Learning method for holdout with an oversampling algorithm, for example the classic SMOTE algorithm. The proposal explained and designed here represents a guideline for the application of oversampling algorithms, but also a brief overview on techniques for overcoming the problem of the unbalanced class in Data Mining.