Semantic Similarity of Healthcare Data

Tags

cosine similarity, healthcare data, semantic similarity

In “Aggregating the syntactic and semantic similarity of healthcare data towards their transformation to HL7 FHIR through ontology matching“, published in the International Journal of Medical Informatics 132:104002 DOI: 10.1016/j.ijmedinf.2019.104002, Kiourtis et al. (2019), address the following objective, and quote:

“Healthcare systems deal with multiple challenges in releasing information from data silos, finding it almost impossible to be implemented, maintained and upgraded, with difficulties ranging in the technical, security and human interaction fields.”

The authors propose an elegant mechanism “that promises healthcare interoperability through the transformation of healthcare data into the corresponding HL7 FHIR structure.”

These are great news! Very cool and practical research that can solve so many problems in the healthcare informatics field.

Many thanks for citing our cosine similarity tutorial as reference 52.

My only reserve with the paper is that early in the article they suggest adding and averaging similarities, which is a mathematically invalid exercise. Distances are arithmetically additive, but similarities (of the same or different kind or source) are not. We can make similarities additive and average them, but not in the arithmetic sense. Other than that, they work is a noble effort.

CUNY Computational Chemistry Tools

Tags

bond order calculator, chemistry tools, computational chemistry, hydrocarbons parser

If you are a chemist, PhD, or student looking for tools in said discipline, you may want to check the City College Chemistry Web Resource Guide, part of the City College of New York Libraries at CUNY. This is an excellent repository of chemistry resources.

This fall, the college redesigned the site so the web address of the computational chemistry section is now https://library.ccny.cuny.edu/chemistry/computational

Happy to know that two of our chemistry tools are still listed there:

The Bond Order Calculator — http://www.minerazzi.com/tools/bond-order/calculator.php Computes bond orders of diatomic species and their ions having up to 20 electrons, including number of bonding and anti-bonding electrons, without using Molecular Orbital Theory (MOT).

The Hydrocarbons Parser http://www.minerazzi.com/tools/hydrocarbons/parser.php — Calculates boiling points and indicates sigma, pi, single, double, and triple bonds for hydrocarbons, again without using MOT.

I wish that more universities follow in the steps of CUNY and be willing to put together similar repositories, I mean computational chemistry tools, for the benefit of their students and faculties.

Why I chose to be a multidisciplinary scientist?

Tags

Binary Distance Calculator, Binary Similarity Calculator, miner, minerazzi, minerazzi tools, Minerazzi Tutorials

When you are a multidisciplinary scientist or teacher, one way of measuring your success is by looking at what students and others in different fields and countries do with the tools and resources you develop. Satisfaction goes all the way up when these help make a difference in their life.

I’m happy to know that in his 2018 PhD thesis “On Enhancing the Security of Time Constrained Mobile ContactlessTransactions” (https://pure.royalholloway.ac.uk/portal/files/33898207/Iakovos_Gurulian_PhD_Thesis.pdf), the author, Iakovos Gurulian from the Information Security Group, Department of Mathematics at the prestigious Royal Holloway, University of London, developed a Python program capable of running our Binary Similarity Calculator (http://www.minerazzi.com/tools/similarity/binary-similarity-calculator.php), which computes 72 different similarity measures. See pages 87-89, tables 4.1 and 4.2, and reference 118 of the thesis.

The Tutorial on Distance and Similarity (http://www.minerazzi.com/tutorials/distance-similarity-tutorial.pdf) was also cited as reference 60.

According to https://www.topuniversities.com/, Royal Holloway ranks 6/10 in London and 291/1000 in the world. Famous for its Founder’s Building, one of the most spectacular university buildings in the world, the College was officially opened by Queen Victoria in 1886.
https://www.topuniversities.com/sites/default/files/royal-holloway-university-of-london.jpg

A Simple Example of Phonetic Similarity vs. Text Similarity

Tags

levenshtein distance, levenshtein similarity, minerazzi tools, phonetic similarity, tools

The followings sound the same so their phonetic similarity is 1.

(a) r u?
(b) ar u?
(c) are u?
(d) r you?
(e) ar you?
(f) are you?

However, the Levenshtein Distance (LD) and Levenshtein Similarity (LS) of (a) with the other strings differ:

LD(a, b) = 1; LS(a, b) = 0.5
LD(a, c) = 2; LS(a, c) = 0.33
LD(a, d) = 2; LS(a, d) = 0.33
LD(a, e) = 3; LS((a, e) = 0.25
LD(a, f) = 4; LS(a, f) = 0.2

Can you find LD and LS results for other possible combinations?

For i = j, LD(i, j) = 0 and LS(i, j) = 1 so you may want to ignore this case.

Note: LD and LS results were computed with our tool at http://www.minerazzi.com/tools/levenshtein/levenshtein-distance-calculator.php

References
http://www.minerazzi.com/tutorials/levenshtein-distance-tutorial.pdf
http://www.minerazzi.com/tutorials/distance-similarity-tutorial.pdf

Zillman’s 2019 Directory of Directories

Tags

directories, Directory of Directories, miner, minerazzi

Zillman’s 2019 Directory of Directories is a handy resource for those interested in finding specialized gateways to the Web.

Glad to know that this summer Minerazzi was included in the Academic/Education section.

(http://columns.virtualprivatelibrary.net/2019_Directory_Of_Directories_June19_column.pdf)

See also http://www.2019directoryofdirectories.com/

Since 2018 we are listed in the Bot and Intelligent Agent Resources category and few other sections, but did not realize that in that one too.

(http://www.zillman.us/minerazzi-your-search-and-mine)

Cool!

More on Perfectoid Spaces

Tags

espacios perfectoides, fractals, Mathematics, perfectoid miner, Perfectoid Spaces

Back in 08-12-2018, we blogged on Peter Scholze’s 2011 thesis on Perfectoid Spaces, one of the hottest topics that is taking Mathematics by storm (https://irthoughts.wordpress.com/2018/08/12/peter-scholze-and-perfectoid-spaces-a-math-genius-among-us/).

An introductory discussion on the topic is available at https://mathoverflow.net/questions/65729/what-are-perfectoid-spaces?rq=1

Possible connections and applications are mentioned at https://irthoughts.wordpress.com/2018/08/23/perfectoid-spaces-arithmetic-geometry-and-quantum-theory/.

The Perfectoid Spaces miner is available since 08-16-2018 (https://irthoughts.wordpress.com/2018/08/16/perfectoid-spaces-miner/).

Resources in spanish were added to the miner yesterday 08-26-2019, about a year later. Better late than never.

Extracting Topic-Specific Wikipedia Links

Tags

miner, minerazzi, wiki links, wikipedia

Below is an illustrative example of a miner for extracting specific links from Wikipedia.

http://www.miislita.com/wikilinks

This miner provides entry points to Wikipedia links relevant to Puerto Rico.

Recrawling a search result discovers new topic-specific links, allowing for further exploration and mining.

Programming Languages Miner

Tags

Python, Programming, PHP, JavaScript, programming languages, computer programming, C++, PERL

Programming Languages Miner (http://www.minerazzi.com/programming-languages/)

Find lists of programming languages, tutorials, and related resources.

Recrawl results to build your own curated collection.

Its news channels include direct access to blog feeds of some of the most popular languages, like C++, JavaScript, PERL, PHP, and PYTHON.

New IANA Miners

Tags

iana, Internet Standards, minerazzi, miners

Here is a series of IANA miners, available now at Minerazzi.com.

IANA Drafts: http://www.minerazzi.com/iana-drafts/

IANA Domains: http://www.minerazzi.com/iana-domains/

IANA Assignments: http://www.minerazzi.com/iana-assignments/

IANA Reports: http://www.minerazzi.com/iana-reports/

Resources relevant to IANA members can also be found through these.

For instance, search for [email]. In the search results page, click the emails tool icon at the top of the results…

How to Lie with Bar Graphs: A Tool for Agents of Misinformation and Fake News

Tags

agents of misinformation, fake news, paid polls, politicians, propaganda, seos

In the article, The Use and Abuse of Logarithmic Axes, Harvey J. Motulsky explained how bar graphs can be used for propaganda and misleading others. The article is at

https://s3.amazonaws.com/cdn.graphpad.com/faq/1910/file/1487logaxes.pdf

He shows the following graph and explains, and quote

Since zero can’t be shown on a log axis, the choice of a starting place is arbitrary. Accordingly, the relative height of two bars is not directly related to their relative values. The graphs below show the same data as the graph above, but with three different choices for where the axis begins. This arbitrary choice influences the relative height of the two bars, amplified in the graph on the left and minimized in the graph on the right.

If the goal is to create propaganda, a bar graph using a logarithmic axis is a great tool, as it lets you either exaggerate differences between groups or minimize them. All you have to do is carefully choose the range of your axis. Don’t create bar graphs using a logarithmic axis if your goal is to honestly show the data.

So now you know. Never trust anyone that uses such tricks (politicians, pollsters, marketers, sales department staff, seos, and the likes). This one is also a great tool for fake news masters.