Introduction to Global Weights with Applications to MySQL

28 Saturday May 2016

Posted by egarcia in Data Mining, IR Tutorials, Machine Learning, News, Programming

Tags

Algorithms, information retrieval, ir, miner, minerazzi, text mining

This is Part 5 of a tutorial series on Term Vector Theory. Several global weight models are discussed and a brief introduction to MySQL implementation of the Vector Space Model presented.

The tutorial is available at

Forward-confirmed reverse DNS (FCrDNS) Tool

24 Tuesday May 2016

Posted by egarcia in Data Mining, Human-Computer Interaction, Internet Engineering, IR Tools, Programming, Queries, Search Engines Architecture Course, Web Mining Course

≈ Leave a comment

Tags

data miners, information security, miner, minerazzi, tools

This tool is available at

http://www.minerazzi.com/tools/fcrdns/lookups.php

The tool allows you to do Forward and Reverse DNS lookups. Given a host name, the tool finds its IP. Conversely, given an IP the tool finds the corresponding host.

Forward DNS lookup resolves a host name to an IP address (A record). The process of reverse resolving an IP address uses the pointer DNS record type (PTR record).

Thus, the tool does Forward-confirmed reverse DNS (FCrDNS) lookups. This is a networking parameter configuration where a given IP address has both forward (name-to-address) and reverse (address-to-name) Domain Name System (DNS) entries that match each other.

Unlike similar tools which do Forward/Reverse DNS lookups on a single host, our tool does lookups on multiple hosts, saving users time and effort.

To use the tool, enter one host name (or IP) per line, ending each line by pressing the Enter key.

Forward DNS lookups are faster than Reverse DNS lookups so for the latter you may want to do a few checks at once.

Depending on DNS server configurations, lookups with or without the www alias can produce dissimilar results. For instance yahoo.com with and without www returns different results.

Applications

Our tool can be used to identify Internet service providers (ISPs) who do not provide properly matching DNS and rDNS records. It can also be used to find shared hosting and, when misconfigured, forwarders information leaks.

FCrDNS verification can also be used for whitelisting purposes because spammers and phishers cannot usually by-pass this verification when they use zombie computers for email spoofing. That is, the reverse DNS might verify, but it will usually be part of another domain than the claimed domain name.

Z-to-P and P-to-Z Transformations

20 Friday May 2016

Posted by egarcia in calculators, Data Mining, Human-Computer Interaction, IR Tools, Marketing Research, Programming, Statistics and Mathematics

≈ Leave a comment

Tags

data miners, Data Mining, minerazzi, statistics, tools

We have developed a new tool that simplifies Z-to-P and P-to-Z Transformations. It is available at

http://minerazzi.com/tools/phi/transformations.php

Unlike similar tools that handle one input score at a time, our tool computes Z-to-P and P-to-Z transformations over an entire set of input scores, saving users time and effort.

The tool facilitates the work of data miners, statisticians, or anyone that need to compute Z and P scores without having to consult Z statistical tables.

It is a great tool for students and teachers interested in Statistics.

What Drives Organic Traffic to Retail Sites?

17 Tuesday May 2016

Posted by egarcia in Big Data, Conferences, Data Mining, Marketing Research, Miscellaneous, News, Queries, Web Mining Course

≈ Leave a comment

Tags

Internet Marketing, minerazzi, news, PPC, search marketing, sem, seo, statistics

If you really want to understand the effectiveness of SEO from a scientific standpoint, read Search Engine Optimization: What Drives Organic Traffic to Retail Sites? from Journal of Economics & Management Strategy, Volume 25, Number 1, Spring 2016, 6–31

I know. The piece is a bit old by Internet standards, but the more one reads it the more one learn. Baye, De Los Santos, & Wildenbeest have done a great research publishing their findings. The study was sponsored by Google, a few years ago.

Actually, the piece was written in 2013, presented at a conference in 2014, and its finding republished in 2016.

It is still very relevant to search marketers and prospective clients.

My favorite piece from their findings is this:

“While previous literature has focused on the role of sponsored links in consumer searches, Michael R. Baye, Babur De los Santos, and Matthijs R. Wildenbeest are the first to explain the drivers of organic (natural) clicks to improve search engine optimization (SEO). In this interview, the authors discuss why investing heavily in brand equity, instead of exclusively on search rankings, is crucial for a retailer’s SEO strategy. ”

and this:

“We also point out that investments that improve site quality and consumer awareness (and more broadly, that enhance an online retailer’s brand equity) are likely to have spillover benefits in other channels that are not accounted for in this or other studies of organic and sponsored search. These benefits include increases in clicks through other online channels (such as price comparison sites), increases in the number of direct visits to a retailer’s website, increases in visits through navigational searches at search engines, and increases in traffic at the retailer’s physical stores. These considerations–coupled with the fact that position is a zero-sum game and thus a retailer is unlikely to obtain a sustainable advantage through direct efforts to improve its ranking–lead us to conclude that brand equity is one of the more important components of retailers’ SEO strategies.”