Lizabeth Wolf

Lizabeth Wolf

Sunday, 16 October 2016 15:14

Why we mustn’t be slaves to the algorithm

The tech craze du jour is machine learning (ML). Billions of dollars of venture capital are being poured into it. All the big tech companies are deep into it. Every computer science student doing a PhD on it is assured of lucrative employment after graduation at his or her pick of technology companies. One of the most popular courses at Stanford is CS229: Machine Learning. Newspapers and magazines extol the wonders of the technology. ML is the magic sauce that enables Amazon to know what you might want to buy next, and Netflix to guess which films might interest you, given your recent viewing history.

To non-geeks, ML is impenetrable, and therefore intimidating. Exotic terminologyabounds: neural networks, backpropagation, random forestsBayesian networks, quadratic classifiers – that sort of thing. Accordingly, a kind of high priesthood has assembled around the technology which, like all priesthoods, tends to patronise anyone who wonders whether this arcane technology might not be, well, entirely good for humanity. “Don’t you worry about a thing, dear,” is the general tone. “We know what we’re doing.”

When I mentioned ML to a classicist friend of mine recently, he replied: “What, exactly, is the machine learning?” That turns out to be the key question. Machine learning, you see, is best understood as a giant computer-powered sausage-making machine. Into the machine is fed a giant helping of data (called a training set) and, after a bit of algorithmic whirring, out comes the sausage – in the shape of a correlation or a pattern that the algorithm has “learned” from the training set.

"The fact the same generic approach works across a range of domains should make you suspicious"

The machine is then fed a new datastream, and on the basis of what it has “learned”, proceeds to emit correlations, recommendations and perhaps even judgments (such as: this person is likely to reoffend if granted parole; or that person should be granted a loan). And because these outputs are computer-generated, they are currently regarded with awe and amazement by bemused citizens who are not privy to the aforesaid algorithmic magic.

It’s time to wean ourselves off this servile cringe. A good place to begin would be to start using everyday metaphors for all this exotic gobbledegook.

Cue Maciej Cegłowski, who describes himself as “a painter and computer guy” who lives in San Francisco and maintains one of the most delightful blogs to be found on the web. Last month, Cegłowski was invited to give a talk at the US Library of Congress in which he proposed a novel metaphor. “Machine learning,” he says, “is like a deep-fat fryer. If you’ve never deep-fried something before, you think to yourself: ‘This is amazing! I bet this would work on anything!’ And it kind of does. In our case, the deep fryer is a toolbox of statistical techniques. The names keep changing – it used to be unsupervised learning, now it’s called big data or deep learning or AI. Next year it will be called something else. But the core ideas don’t change. You train a computer on lots of data, and it learns to recognise structure.”

“But,” continues Cegłowski, “the fact that the same generic approach works across a wide range of domains should make you suspicious about how much insight it’s adding. In any deep-frying situation, a good question to ask is: what is this stuff being fried in?”

The cooking oil, in the case of machine learning, is the data used for training. If the data is contaminated – by error, selectivity or bias – so too will be the patternslearned by the software.

And of course, the ML priesthood knows that, so the more conscientious practitioners go to considerable lengths to try to detect and correct for biased results in applications of the technology. For an increasing number of ML applications, though, the training sets are just huge collections of everyday conversations – culled, for example, from social media. That sounds OK: after all, ordinary speech is just that. But a remarkable piece of research by AI researchers at Princeton and the University of Bath reveals that even everyday speech has embedded biases of which most of us are unaware. “Language itself contains recoverable and accurate imprints of our historic biases,” they write, “whether these are morally neutral as towards insects or flowers, problematic as towards race or gender, or even simply veridical, reflecting the status quo for the distribution of gender with respect to careers or first names.” And of course these hidden biases are inevitably captured by machine learning.

I suspect that Wittgenstein would have loved this research: it confirms his belief that the meaning of a word is not to be found in some abstract definition, but in its use in everyday language. Maybe ML geeks should read his Tractatus.

Source: theguardian.com

I found your original ad in the Woman’s World Magazine and have been searching all through the Pinterest ads for work from home jobs. There are so many to go through and some want money and others I am not sure are real jobs but Holly, I LOVE research and would like to start a business doing research, but have no idea how to get started. I thought I would ask if you had any ideas since you are seeking ways of possible employment for people.

I hope you will have some ideas and share them with me as am very eager to get started.

Sincerely, Judy L.

Hi Judy,

Internet Research is a great way for you to work from home! In fact, there are lots of companies who hire freelancers to do their research for them — you just need to know where to look. Law firms, marketing departments, insurance companies, healthcare providers, political groups, even writers, publishers, and college students need researchers. If you have background experience in one of the aforementioned areas, I suggest starting there first, as you’ll have a leg up on your competition.

Here’s where to find the gigs and how to get started.

Online Job Boards.

Indeed is one of my favorite sites to use when searching for telecommuting jobs, here’s why. It’s an aggregated job board, which means their platform pulls job listings from all the other major job board sites. This means you save tons of time, because you don’t have to search each job site individually. While the jobs are not screened for legitimacy, Indeed does requires that all listings contain a company name, location, job title, and complete job description.

To find Internet Research jobs just use the keywords, “work at home” and “internet research”.

FlexJobs is an online job board that caters to flexible working arrangements. Simply use their job search function and enter “Internet Research” or “Internet Researcher”. This will take you to a page that lists all the jobs in this category. While it does cost a small fee to join, it’s well worth it. All jobs are hand screened, so you know that you’re dealing with legit opportunities.

Upwork is freelance jobs board sites that cater to freelancers. Simply register for an account (it’s free), search for Internet Research positions, and submit your bid. If your bid is accepted, you’ll complete the work and get paid through the platform, minus a small fee (usually 10%). While many freelancers complain of the low rates on these sites — it can be a great way for you to establish yourself in the field, and to work with repeat clients.

Also, check out this article from Copyhackers, on how Danny Margulies was able to earn six figures from Elance. He really has a neat strategy!

Answer and Research Questions.

There are a bunch of sites that hire independent contractors to answer questions on various topics. So if you have experience in a certain area that will be extremely helpful when you’re applying. With these opportunities you’re not only researching answers for people, you’re also writing the answers online — so you’ll need to write well.

Experts 123 – Is a revenue sharing platform, so the more popular your answer they more you’ll get paid. Payments are made via PayPal.

Just Answer – Pays 20% – 50% of what the person is willing to pay for the answer (if it’s accepted). Payments are made via PayPal. Has an A+ rating with the Better Business Bureau.

Small Biz Advice – Browse questions and make a bid. Payments are made bi-monthly via PayPal, minus a 5% commission fee.

Web Answers – Incorporates Google Adsense, so you get paid based on how much traffic your answer receives.

Wonder – Hires internet researchers to gather product info, explain trends, gather stats, and determine the size of a market. Researchers are paid per assignment.

Launch Your Own Internet Research Business.

1. Educate yourself on the topic of Internet Research.

There aren’t a lot of books on this topic, but here are a couple:

  • Start & Run an Internet Research Business (Start & Run Business Series)
  • Secrets of Becoming an Internet Research Specialist: How to Surf the Web for Freedom and Profit

2. Get all of your legal and administrative ducks in a row.

Choose your business structure, fill out your DBA (Doing Business As) form, set up a PayPal account, so that you can accept credit card and online payments. You’ll also need to consider things like health insurance, retirement, and taxes. This article explains a lot of what you need to know.

3. Set up shop.

Next set up your website, social media profiles, and email marketing account. These three items will help you market and advertise your new business.

  • 5 Free Email Marketing Services
  • Building a Website

4. Get clients.

There’s a variety of ways to do this, but first start with the job boards I mentioned in this post.

Here are some other ways in which you can get clients.

  • The Way to Find More Clients and Make More Money
  • How to Get More Business Clients and Customers

Good luck with your Internet Research Business and keep me posted on your progress!


It turns out those rumors last week were accurate. Microblogging site Twitter has acquired the Summize search engine, Twitter co-founder Evan Williams confirmed to me. The size of the transaction is not being disclosed, although the transaction price was paid “mostly in stock.”

Twitter has also hired 5 of the 6 Summize employees. Founder and CEO Jay Verdy will move on to a new project.

The five Summize employees joining Twitter are all engineers, adding to the twelve engineers that currently work at Twitter. Summize CTO Greg Pass will become Twitter’s top tech guy as Director of Engineering and Ops.

The deal has been discussed for some time, Twitter co-founder Evan Williams told me earlier. The companies had a term sheet in place when Twitter partnered with Summize in June to help them keep the Twitter platform stable during the Apple iPhone 3G Steve Jobs keynote. The deal was closed in the last week.

John Borthwick, a partner at Summize investor Betaworks, was also an investor in Pyra Labs (Blogger.com), which Evan Williams co-founded in 1999. In other words, the companies were already kissing cousins.

I spoke with Williams over the weekend at Foo Camp about the transaction and other Twitter issues. The video is below. We’ll post a full transcript later today.


Days after Verizon announces it will acquire Yahoo, Yahoo is testing a new search bar at the top with a missing logo at the left.

Less than a week after Verizon announced it will buy Yahoo, Yahoo is testing a new search interface.

The new interface removes the Yahoo logo from the top left of the screen, makes it a bit smaller and moves it to the right side of the screen. This leaves an open white space at the top left, maybe potentially for the Verizon logo? Probably not. It would be too soon for Yahoo to test Verizon branding on their properties, I would think.

With this user interface change, you will also see a new box at the top right that opens up a Yahoo services navigator. Also, the search button is blue, instead of the Yahoo purple.

Here is a screen shot from all google testing blog.

Yahoo tests a new SERP interface


How can a business stand out in local SEO with no physical premise? Is it even possible?

Local search is very competitive and it becomes even more of a challenge to compete with other businesses when yours has no physical premise.

Sam Nemzer from Distilled shared with us some useful tips at MozTalk on how to use local SEO for a business with no local pack.

Here are the five tactics that Sam Nemzer suggests:

1) Put user-focused content on category pages

A category page should focus on content and the more you add, the better for the SERPs. In order to add content, you can:

  • Pull in third party data to immediately gain access to useful and relevant content (Facebook, Twitter, Flickr, Wikipedia, Google Maps, Ticketmaster, TripAdvisor, etc)
  • Use proprietary data that may be helpful for your business (this also makes it easier to obtain featured snippets)
  • Find questions to answer. Sam Nemzer suggested the tool “Answer the Public”, a free visual keyword research that allows you to find the right questions for the specific keywords you want to use.

answer the public

By the time you’ve found the answers from the particular keywords, it’s time to answer different questions on your own:

“How can you use this content on your site?”

“Is this still relevant to your business?”

“Have I relied too much on auto-generated content?

Remember, Google is not keen on pages that focus on automatically generated content, so make sure you don’t overuse it.

2) Get the level of granularity right

It’s important to start by researching your analytics data, to understand what gets traffic and how you can increase the content on your site.

Have a closer look at the keywords and see what people search in your industry.

Not every site needs the same details and the example below shows how property listings differentiate from job sites.

MozTalk - local SEO

Finally, it may be useful to analyse your keyword data and ask yourself: “Do you need location pages at all?”

How can your business benefit from them? By the time you’re able to answer this question, you’re ready to continue to more practical tips.

3) Sidestep local packs by targeting high in the funnel

In order to avoid the high competition, it may be useful to avoid focusing on the most popular local packs and find the right content gaps to exploit.

How about aiming for different targeting where there’s less competition and more opportunities to stand out?

Retargeting may be useful to reduce CPC, but this also means that the conversion may not be immediate, although this also means that you can build recognition and serve as a knowledge provider in your industry.


Friday, 20 May 2016 11:05

Research Tools

The Internet can be a researcher's dream come true. By browsing the Internet, much as you would browse the shelves of a library, you can access information on seemingly limitless topics. In addition, web-based catalogs are available in many libraries to assist researchers in locating printed books, journals, government documents, and other materials.


Possibly the biggest obstacle facing researchers on the Internet is how to effectively and efficiently access the vast amount of information available with the simple click of the mouse. With the Internet's potential as a research tool, teachers must instruct and guide their students on manageable strategies for sorting through the abundance of information. The search for reliable resources can be both overwhelming and frustrating if students are left on their own in their initial search. A few simple guidelines can make conducting research more manageable, reliable, and fun.



The research process


Lessons and projects should be designed so that research time on the Web can be maximized in terms of efficiency. This may mean gathering necessary information beforehand, having students work in groups, or focusing on whole-class projects.

Barron and Ivers (1996) outlined the following cycle for online research projects.








Step 1: Questioning --- Before going on the Internet, students should structure their questions.
Step 2: Planning --- Students should develop a search strategy with a list of sites to investigate.
Step 3: Gathering --- Students use the Web to collect and gather information.
Step 4: Sorting & Sifting --- Students analyze and categorize the data they gathered on the Web.
Step 5: Synthesizing --- Students integrate the information into the lesson, and draw conclusions.
Step 6: Evaluating --- Students assess the results, and if necessary, begin the process again.



Searching the Web


There are billions of pages of information on the World Wide Web, and finding relevant and reliable information can be a challenge. Search engines are powerful tools that index millions of web sites. When entering a keyword into a search engine, you will receive a list with the number of hits or results and links to the related sites. The number of hits you receive may vary a great deal among different search engines. Some engines search only the titles of the web sites, and others search the full text.


Techniques for using the different search tools vary. For best results, read the search tips or hints that are provided at each search site. Also, note that some of the search engines do not allow Boolean searches that combine words with the logical connectors of AND, OR, or NOT.


Common commands for search engines include:

Quotation Marks ( " )
Using quotation marks will help to find specific phrases involving more than one word. For example: "Martin Luther King"
Addition Sign ( + )
Adding a + sign before a word means that it MUST be included in each site listed. For example: + Florida + taxes
Subtraction Sign ( - )
Adding a - sign before a word means that it will NOT appear in the sites listed. For example: + Washington -DC
Asterisks ( * )
Asterisks can be used for wild-cards in some search engines. For example: Mexic* will look for Mexico, Mexican, Mexicali, etc.



Search engine capabilities


Search engines are rated by the size of their index. Large engines such as Google are good tools to use when searching for obscure information, but one drawback to an extensive index is the overwhelming number of results on more general topics. If this is the case, it might be better to use a search engine with a directory structure such as Yahoo.


Many search engines provide directory-listing search tools such as yellow pages, white pages, and email addresses. In addition, many allow you to personalize their site to your needs. For example, you might want to set the attributes of the page to show educational news headlines and your favorite teacher resource links. In the preferences of your web browser, you can then set this page as your home start-up page.



Search engines especially for children


Search engines designed for younger students are useful tools for the classroom. They screen for inappropriate material and provide appropriate sites for students on topics related to educational and entertainment purposes. Using these sites helps to narrow the scope of hits on a search inquiry. As a result, the student will spend less time reading irrelevant material.


Although some search engines allow you to turn on filters to help filter out adult content, they are not always thorough or accurate. There are several good search engines that are specifically designed for the younger audience, such as Ask Jeeves and Yahooligans.



Evaluating Internet sources


Students often uncritically accept information they see in print or on computer screens. Students should be encouraged to carefully evaluate sources found on the Internet. The evaluation tool (below) will help students analyze web resources in terms of accuracy, authority, objectivity, timeliness, and coverage. Consideration of these factors will weed out many of the inaccurate or trivial sites students may encounter.



Analyzing web resources


Answer the following questions to evaluate web resources.
Are sources listed for the facts?
Can information be verified through another source?
Has the site been edited for grammar, spelling, etc.?



Is the publisher reputable?
Is the sponsorship clear?
Is a phone number or postal address available?
Is there a link to the sponsoring organization?
Is the author qualified to write on this topic?



Does the sponsor have commercial interests?
Is advertising included on the page?
Are there obvious biases?



Is a publication date indicated?
Is there a date for the last update?
Is the topic one that does not change frequently?


Are the topics covered in depth?
Does the content appear to be complete?



Setting bookmarks on the Web


Browsers such as Safari, Firefox, and Internet Explorer provide a way to create a list of your favorite sites that you can access with a click of the mouse. The procedure for creating a list of sites is an easy and powerful tool for web use. When you find a web page that you want to bookmark, simply select the "Add Bookmark" or "Add Favorite" option from the menu bar. To return to the site at a later time, choose the name from the bookmark or favorite list, and you will immediately access the site. You can organize your bookmarks into file folders and can save them on a disk to transfer and use on other computers.



Copyright issues


Teachers and students have a somewhat flexible, but not unlimited, copyright privilege under the "fair use clause" of the U.S. Copyright Act. "Fair use" is the means by which educators of non-profit educational institutions may use copyrighted works without seeking permission or making payment to the author or publisher. Teachers and students are also protected to some extent by the Digital Millennium Copyright Act, which went into effect in October 1998. Under current guidelines, teachers and students are able to make limited use of copyrighted materials for instructional purposes.


Currently, copyright law as it relates to the Internet is vague and being challenged and rewritten on an ongoing basis. However, the guidelines of the "fair use clause" can be applied to Internet use in the classroom. Although classroom use allows teachers and students to be creative, you must also be extremely careful. Teachers and students should realize that all materials found on the Internet are protected by the same copyright laws as printed materials. Copyright protects "original works of authorship" that are in a tangible form of expression.


Copyrightable works include the following categories:


literary works
musical works, including any accompanying words
dramatic works, including any accompanying music, pantomimes, and choreographic works
pictorial, graphic, and sculptural works
motion pictures and other audiovisual works
sound recordings
architectural works
These categories should be viewed broadly. For example, computer programs and most "compilations" may be registered as "literary works"; maps and architectural plans may be registered as "pictorial, graphic, and sculptural works."


Important questions to ask


What is the purpose for using the material?
Who is the audience?
How widely will the material be distributed?
Will the material be reproduced?
It is allowable under copyright guidelines to use copyrighted materials for class assignments. Check specific guidelines for length of time the material can be kept up on a web site.


When in doubt, ask.

If you and your students find a graphic or portion of a text on the Internet that you want to utilize in a class project, locate the source of the web site and email them to ask permission for use of their graphic or text. Many web site designers are happy for you to "borrow" their graphics and words. Some ask that you give them credit and others do not. Although your students may be too young to comprehend copyright law, they can understand the concept of respecting someone else's property.


It is advisable for school sites to have an online service provider or an "agent" who can act as a filter on copyright issues. The agent would be the person someone would notify if they found a copyright violation on a student or school web site. In most cases, you are simply asked to remove the offending copyright violation.


For more information on fair use guidelines for educational multimedia, go to the
Multimedia Fair Use Guidelines (CCMC) web site.


Copyright discussions with students may include:


Does copyright apply to student web pages? Any original work of authorship, whether created by a student, teacher, or professional is protected by the copyright laws. An original piece of work does not need to possess or display a copyright to be protected under the copyright laws.
May students "borrow" art, sound, animation, etc., from others' web pages? Resources (such as graphics and sound files) from most web sites are copyright protected and require permission to use, but the resources at some web sites are advertised as "free" for use. These web sites may require that credit is given to the original source of the materials.



Student activity: Finding a Needle in Cyberspace


Using the major search engines on the Web, find the best way to look for a needle. Fill out the following chart, noting the number of hits you receive in each of the search engines for the word needle and the phrase "Space Needle." Then, answer the questions at the bottom of the page.

Search Engine Search for: needle Search for: "Space Needle"


Which search engine would be the best if you were looking for something very obscure?
Did searching for "Space Needle" always result in more hits or less hits than searching for needle? Why?
Which search engine seemed to display the result fastest?
Try another search. This time, look for sites that contain all of these words: needle, sleeping, and beauty. (Hint: On many of the search engines you can specify that certain words MUST be included by adding a + in front of the word: +needle +sleeping +beauty.)


Source:  http://fcit.usf.edu/internet/chap5/chap5.htm












Every day, Internet users are literally bombarded with data, making it impossible at times to complete work tasks or home tasks due to technological distraction. An article published last year by the Harvard Business Review further backs this overload of content problem with figures. According to Harvard Business Review, we produce more Internet data every second compared to the entire Internet’s storage of date 20 years ago. That is astounding, so it’s no wonder that we’re now dealing with this issue.

History of Information Overload

This is actually not a new phenomenon. In fact, this issue dates back to “movable type” and printed matter. Further technologies later exasperated information overload by allowing instant access to data through digitalized content. The barriers of printing presses were removed, allowing for the instant publishing of new content.

And as many content creators and future content creators caught onto just how invaluable the Internet actually is for the delivering of new content to users, the competition began. Now, with so many bloggers and businesses scrambling to create and publish new content, we have bombarded ourselves with a deluge of data and content. 

Problems Associated with Information Overload

Some common problems with this phenomenon include: 

  • Feeling overwhelmed
  • Feeling distracted
  • Problems with overall wellbeing
  • Problems with decision-making
  • Interruptions of work tasks
  • Loss of revenue for businesses
  • Interruptions of home tasks

According to recent research of this problem, the above effects have been linked to an overload of information. Relating to business, an overload of information is known to lead to a reduction in productivity and difficulty making decisions. One study found that it took an average of 25 minutes for many workers to return to work following an e-mail or RSS feed interruption.

However, with that said, many users and workers recognize the issue at hand, but are at a loss as to how to overcome this problem. Because many feel that keeping up with the latest information and research is also an integral part of their job. That’s where recovery techniques come into play to help overcome an overload of information. Learn how to overcome this problem with tips and believe it or not, more technology.

How to Recover from Information Overload

Tips for the user bombarded with too much information:

  • Improve self-control of the impulse to check your e-mail while you’re busy working on something else.
  • Prioritize data.
  • Set time limits for viewing new data.
  • Set schedules for data access that will not interfere with your work schedule.
  • Allow the latest technologies available to help you do all of the above to help you improve self-control and prioritize data.

Content Marketing Strategies to Overcome Information Overload

Activity Streams:  By subscribing to Activity Streams, you could help to reduce an overload of information, because when you receive a list of recent activities to a user or business that you have subscribed to, you then have the choice to only access the content that interests you most. Instead of feeling bombarded with everything at once, the content has been condensed into a neat and tidy list for you to choose from.

Filter Incoming Messages:  Filter incoming messages by priority. For instance, with Google you can now filter messages into various folders such as Inbox, Promotions, Work, Spam, etc.

Focus and Self-Discipline:  Focus only on content and users that truly interest you. Why subscribe to data that you know you won’t have time to view or really have no interest in viewing? Only subscribe to content that you truly feel brings value to your world and then discipline yourself as to how much time you allow for viewing this data and set times that work best with your busy schedule to view data. This is a great way to avoid becoming overwhelmed.

When Producing New Content:  Keep your content fresh, unique, valuable and relevant.

Conclusion to Information Overload

At first, it may feel like this is hard to do, if you feel almost addicted to content. But in the end, once you start implementing these easy steps, you’ll realize just how easy they really are, you’ll feel better overall and you’ll notice a significant difference in completing tasks online and offline.

Source :

Page 2 of 2

airs logo

Association of Internet Research Specialists is the world's leading community for the Internet Research Specialist and provide a Unified Platform that delivers, Education, Training and Certification for Online Research.

Get Exclusive Research Tips in Your Inbox

Receive Great tips via email, enter your email to Subscribe.

Follow Us on Social Media