New paper on normalization of inconsistency indicators – to appear in Journal of Approx Reason

Authors: Waldemar W. Koczkodaj, Dominik Strzalka, Jean-Pierr Magnot, Jiri Mazurek, James Peters, Michael Soltys, Jacek Szybowski, Arturo Tozzi, Hojjat Rakhshani

Title: On normalization of inconsistency indicators in pairwise comparisons

Abstract: In this study, we provide mathematical and practice-driven justification for using [0, 1] normalization of inconsistency indicators in pairwise comparisons. The need for normalization, as well as problems with the lack of normalization, are presented. A new type of paradox of infinity is described.

Accepted for publication in the International Journal of Approximate Reasoning, April 2017.

A new paper on normalization of inconsistency indicators

A new paper: On normalization of inconsistency indicators in pairwise comparisons, by W.W. Koczkodaj, J.-P. Magnot, J. Mazurek, J.F. Peters, H. Rakhshani, M. Soltys, D. Strzałka, J. Szybowski and A. Tozzi.

Abstract: In this study, we provide mathematical and practice-driven justification for using [0,1] normalization of inconsistency indicators in pairwise comparisons. The need for normalization, as well as problems with the lack of normalization, are presented. A new type of paradox of infinity is described.

The paper can be found here: https://arxiv.org/abs/1702.07205v2

CI press release re my forthcoming SAME talk on cybersecurity

CSUCI Cyber-security expert to speak at on-campus engineering convention.

Camarillo, Calif., Feb. 20, 2017—Whether it’s personal information, medical records, national security or election results, computer hacking is a rising national and global concern.

CSU Channel Islands (CSUCI) Computer Science Chair and Professor Michael Soltys, Ph.D., will share his cybersecurity expertise to an audience of professional engineers from 11 a.m. to 2 p.m. on March 23 in the Grand Salon.

“Our society is under constant cyber-threat, as our infrastructure, our economy, and our privacy, depend on secure IT systems,” Soltys said. “My talk will consider the major threats, and present examples of how hackers attack our systems.”

Sponsored by the nonprofit Oxnard Ventura Post of the Society of American Military Engineers, the presentation is geared toward an audience with a high degree of computer expertise, so Soltys plans to share cybersecurity best practices.

“I plan to give more of a technical talk from the engineering point of view,” Soltys said. “How to write code that is more defended. I plan to show techniques hackers use to get into systems.”

One of the principal causes of cyber-vulnerability is faulty software, a problem Soltys addresses in a textbook on algorithms he wrote for software engineers.

Aside from his teaching at CSUCI, Soltys also acts as Director of IT Security for Executek International where he specializes in forensic work.

The public is welcome at the presentation, which is on campus at One University Drive in Camarillo. Cost is $30 a person for lunch.

Follow the directional signage to Parking Lots A-4 and A-11, then follow “walk this way” signage to the Grand Salon.

To register for the presentation, click on:

http://www.same.org/Oxnard-Ventura

 

Indeterminates in London

Today my masters student Joel Helling presented our joint work, Constructing an Indeterminate String from its Associated Graph,  co-authored with P.J. Ryan, W.F. Smyth, at the LSD & LAW 2017 conference. This work started with the visit of W.F. Smyth at CSU Channel Islands in March 2016 – see here, when Joel, Bill and I started working on an initial algorithm designed by Joel; this algorithm computed a labeling for a graph such that any two nodes that shared an edge would also share a label, and would not share a label otherwise. Of course, the easy way to do this is to assign each edge a unique label; but there are better ways of doing it, in the sense that fewer labels may be required (for example, imagine a clique: then a single label would do it). This problem is directly related to indeterminate strings; strings that arise in bio-informatics, where certain positions may consist of several possible symbols. For example, {a,b}ab{a,b,c}{b,c}a, where the first position is not determined (it could be a or b), the second position is a, the third b, the fourth is again not determined (it could be a or b or c), etc. Such strings are well suited to represent genetic
information which often contains “noise”. The paper is now accepted for publication in Elsevier’s journal of Theoretical Computer Science,  as well as presented by Joel at the conference.

I would like to express my gratitude to the British Royal Society for awarding us an exchange grant, CSU Channel Islands / King’s College London, which financed the trip of Joel Helling.

CI Facebook post.

Neerja Mhaskar and I have a paper presented at PSC 2015

We (Neerja Mhaskar and Michael Soltys) had our paper presented at the Prague Stringology Conference; the paper introduces a new formal framework for Stringology, which consists of a three-sorted logical theory S designed to capture the combinatorial reasoning about finite words:

And these are the slides:

Square-free strings over alphabet lists

In a new paper, Square-free strings over alphabet lists, my PhD student Neerja Mhaskar and I, solve an open problem that was posed in A new approach to non repetitive sequences, by Jaroslaw Grytczuk, Jakub Kozik, and Pitor Micek, in arXiv:1103.3809, December 2010.

The problem can be stated as follows: Given an alphabet list $L=L_1,\ldots,L_n$, where $|L_i|=3$ and $0 \leq i \leq n$, can we always find a square-free string, $W=W_1W_2 \ldots W_n$, where $W_i\in L_i$? We show that this is indeed the case. We do so using an “offending suffix” characterization of forced repetitions, and a counting, non-constructive, technique. We discuss future directions related to finding a constructive solution, namely a polytime algorithm for generating square-free words over such lists.

Our paper will be presented and published in the 26th International Workshop on Combinatorial Algorithms (IWOCA), Verona, Italy, October 2015.

Very happy to have received a grant from the Royal Society to travel between CI and King’s College London

I am very pleased to have received a grant from the British Royal Society to travel between CI and King’s College London over the next two years. This grant will enable collaboration in the field of String Algorithms between CI and King’s College, and it is a joint effort with Maxime Crochemore and Costas Iliopoulos.

The Royal Society is a Fellowship of the world’s most eminent scientists and is the oldest scientific academy in continuous existence. We aim to expand the frontiers of knowledge by championing the development and use of science, mathematics, engineering and medicine for the benefit of humanity and the good of the planet.

A new paper: String Shuffle, with Neerja Mhaskar

Title: String Shuffle: Circuits and Graphs

Authors: Neerja Mhaskar and Michael Soltys

Abstract: We show that shuffle, the problem of determining whether a string w can be composed from an order preserving shuffle of strings x and y, is not in AC0, but it is in AC1. The fact that shuffle is not in AC0 is shown by a reduction of parity to shuffle and invoking the seminal result of Furst et al., while the fact that it is in AC1 is implicit in the results of Mansfield. Together, the two results provide a lower and upper bound on the complexity of this combinatorial problem. We also explore an interesting relationship between graphs and the shuffle problem, namely what types of graphs can be represented with strings exhibiting the anti-Monge condition.

Source: