Group items tagged

Adam Rogers, Science, Wired, 8.6.15, writes about how the Google tanking algorithm of positive and negative stories on the candidates could affect major elections 25% of the timer. This is the tyranny of the algorithm. They tested the impact in mock voter labs before elections in Australia and India where the impact of feeding positive stories about a candidate first shaped voters decisions between 24 and 72 percent of the time with certain voter groups. Voters in towns in the US that watch a local a Fox channel vote more conservatively because of recency and placement issues. While the numbers in real live do not add up to the impact achieved in the test research, when elections are decided by 1 or 2 percentage points, it's enough to turn the tide in favor of a candidate.

Harvard Business Review's top trends identified by using 5 questions that determine if these are indeed a trend or not. Top trends for 2015 are: Deep learning (machine learning); Smart virtual personal assistants; Uber's monetization of downtime and the offer for those needing employment to work. Uber-like businesses such as grocery delivery, massage services, dry cleaning and laundry, etc. will take off; Oversight for Algorithms - ethics of how algorithms can be used especially when programmers add subjective judgments to algorithms causing false answers; Data privacy - dealing with ongoing breaches. The public does not blame hackers but blames business for not taking measures to combat hackers; Block chain technology is a transactional database that is shared by everyone participating in bitcoin's digital system. Block chain systems may become a universal platform for anything needing a signature or authentication.

Harvard Business Review's top trends identified by using 5 questions that determine if these are indeed a trend or not. Top trends for 2015 are: Deep learning (machine learning); Smart virtual personal assistants; Uber's monetization of downtime and the offer for those needing employment to work. Uber-like businesses such as grocery delivery, massage services, dry cleaning and laundry, etc. will take off; Oversight for Algorithms - ethics of how algorithms can be used especially when programmers add subjective judgments to algorithms causing false answers; Data privacy - dealing with ongoing breaches. The public does not blame hackers but blames business for not taking measures to combat hackers; Block chain technology is a transactional database that is shared by everyone participating in bitcoin's digital system. Block chain systems may become a universal platform for anything needing a signature or authentication. 

blog post by Quentin Hardy, NYT, on how new companies developing algorithms are using them to loan money to people who are better risks than their financial circumstances might suggest, track high performers in sales jobs to find the indicators of their success for export and use by other employees, etc. July 26, 2015

article by Claire Cain Miller, New York Times, printed in Tampa Bay Times on 8.14.15

interesting assessment of algorithmic limitations in providing us with access to the online information we really want, Jarno M. Koponen, June 25, 2015. reminds of comment I saw "tyranny of algorithms"

how algorithms and humans act together now for search engines and finding info and categorizing info.

Interesting article on algorithms and humans working together to retrieve and catagorize as well as provide search engine information.

Beginning of a series on big data, algorithms, and some consequences by Martin Muller, Marcel Rosenbach, and Thomas Schulz in Spiegel Online International, May 17, 2013

algorithm cartoon by Rob Cottingham, August 2015

article by Kelly Jackson Higgins at Dark Reading.com on what's happening with the sale of online data collected legally, but not necessarily analyzed accurately or sold ethically. November 5, 2014

how they are microseconds ahead to make financial moves. Talk by Kevin Slavin, How Algorithms Shape Our World.

part of a series in Spiegel Online--this one focuses on building algorithms

blog post by David Dietrich, June 25, 2013, director of technical marketing for the big data solutions group at EMC, founder of an initiatve at MIT, big data@CSAIL on how privacy is sacrificed online because of big data collection and algorithms that can figure out profiles based on seemingly unrelated information.

" 5 Reasons Why Job Boards Aren't As Effective Anymore Don Goodman November 24, 2015 Job Search At one time, job boards were the way to go for job seekers. It's where you could post your resume for employers and recruiters to view, and apply to job openings. But today, it's a different story. Related: Reactive Vs. Proactive Job Search Strategies Job boards are simply not as effective anymore since there are social media outlets like LinkedIn, Facebook, and Twitter where you can pretty much network your way to the right contacts. The fact is, job boards have a 2-4% effectiveness rate whereas networking has over a 50% effectiveness rate." More than 85% of employers use Applicant Tracking Systems (ATS) to review and rank resumes according to skills, experience, keywords. Companies use internal algorithms, so out of an average of 400 resumes using these measures results in only 10 - 20 even looked at. Most hiring managers and recruiters use Linkedin first. Job board resumes are still viewed, but chances are the info is outdated so relying on Linkedin makes sense for recruiters. Niche job boards are worth going to, but to be more productive tie into direct networking through the right contacts.

" 5 Reasons Why Job Boards Aren't As Effective Anymore Don Goodman November 24, 2015 Job Search At one time, job boards were the way to go for job seekers. It's where you could post your resume for employers and recruiters to view, and apply to job openings. But today, it's a different story. Related: Reactive Vs. Proactive Job Search Strategies Job boards are simply not as effective anymore since there are social media outlets like LinkedIn, Facebook, and Twitter where you can pretty much network your way to the right contacts. The fact is, job boards have a 2-4% effectiveness rate whereas networking has over a 50% effectiveness rate." More than 85% of employers use Applicant Tracking Systems (ATS) to review and rank resumes according to skills, experience, keywords. Companies use internal algorithms, so out of an average of 400 resumes using these measures results in only 10 - 20 even looked at. Most hiring managers and recruiters use Linkedin first. Job board resumes are still viewed, but chances are the info is outdated so relying on Linkedin makes sense for recruiters. Niche job boards are worth going to, but to be more productive tie into direct networking through the right contacts.

story of how algorithms and big data sealed the deal for Trump

41 minutes into the panel--they talk about Facebook algorithm and lack of privacy

Blog, June 19, 2013, Scoop.it, on how Most Popular leads into reading more in a webiste can be quite distorted to direct reads. Advocates for human curation. "The solution: Human curation. We need to provide readers with lists and collections of content that are popular within a certain field because they are genuinely popular to those who know about or are interested in that field. Most Popular lists on websites are generated by algorithms, making them extremely easy to game (by clicking the same content repeatedly, for instance) which in turn leads readers to believe they've found good content when they oftentimes haven't. It's the job of the curator, as a human, to let other humans know what they (the curator) are finding to be the best pieces of content on a topic - as someone who knows about a specific subject or interest."

"A comment on Joho's (Dave Weinberger) blog post--a memoir of blogging--bySuw Charman-Anderson, January 9, 2014. Excerpt: " I wonder too if my lack of blog writing is related to a lack of blog reading. My RSS reader became so clogged that I feared it, wouldn't open it, and ultimately, abandoned it. And then Twitter and now Zite arrived to provide me with random rewards for clicking and swiping, showing me stuff that I had no idea I wanted to read. Instead of following the writings of a small cadre of smart, lovely people whom I am proud to call my friends, I read random crap off the internet that some algorithm thinks I might be interested in, or that is recommended by the people I follow on Twitter. That may or may not be a good thing. We were all aware of the problems of homophily, and the random clickage does help combat that. But the problem with not following people's blogs closely is that there's no conversation anymore. My blogs used to host great conversations, and I would happily engage in fascinating discussions on other people's sites. You can't do that so easily with Twitter, and Facebook. Indeed, most of my interactions on Facebook, which are scarce as I loathe it, end up being pointless arguments with friends-of-friends who turn out to be idiots. I'd love to see a resurgence in blogging. I think, personally, I need to delete Zite from my ipad and find a good RSS reader so I can follow the blogs of those people that I really care about. Not the worthy blogs I ought to read, but the works of people who matter to me. And then I need to get back to commenting, like this, because there's nothing more encouraging than finding out that people care about what you write, that people appreciate it. And David, I really do appreciate your writing - you're as inspiring and fascinating now as you were back in 2001! "

article by Martin Dewhurst and Paul Willmott, September 2014 on new leadership skills required in age of new information technologies Machines force executives and senior leaders to: 1. open up their companies through crowdsourcing and social platforms within and across organizational boundaries 2. create data sets worthy of the most intelligent machines 3. "let go" in ways that run counter to a century of OD 4. executives...able to make the biggest difference through the human touch. ...questions they frame, their vigor in attaching exceptional circumstances highlighted by increasingly intelligent algorithms ... tolerating ambiguity and focusing on the "softer" side of management to engage the organization and build its capacity for self-renewal. 5. turbocharged data-analytics strategy, a new top-team mind-set, fresh talent approaches, and a concerted effort to break down information silos...transcend number crunching..."weak signals" from social media and other sources also contain powerful insights and should be part of the data-creation process. 6. ...early movers will probably gain insights of unstructured data, such as email discussions between representatives or discussion threads in social media. 7. ...dashboards don't create themselves. Senior executives must find and set the software parameters needed to determine, for instance, which data gets prioritized and which gets flagged for escalation. 8. ...odds of sinking under the weight of even quite valuable insights grow as well. Answer: democratizing it: encouraging and expecting the organization to manage itself without bringing decisions upward. ...business units and functions will be able to make more and better decisions on their own. 9. 8 will happen even as the CEO begins to morph into a "chief experimentation officer," who draws from acute observance of early signals to bolster a company's ability to experiment at scale. 10. need to "let go" will be more significant and the discomfort of s

Amazing blog post by Kevin Hodgson (dogtrax), July 22, 2015, on edublogs on how the internet, once nodes and networks, and little hierarchy, is changing and is becoming owned by the algorithm-exercising interests of Facebook, Twitter, etc. Instead of a flooded plain, we have streams controlled by specific owners, in which we can not escape.