Online and Social Media Analysis: Research and Government Customers #InternetLibrarian

Alison Ounanian, Stacey Galik, MITRE

Think tank for the federal government.

Governments using social media during emergencies. Project for Dept. of Homeland Security.
* Current events
* Faster updates
* Analysis of possible future events

Used a social media tool called Crimson Hexagon.

Project was called DHS Smart Cities. Studied the area around Hampton Roads, Va. Vulnerable to sea level rise and flooding. Several military bases in the area. Retrieved relevant social media posts from several sites for Nov. 2014-Nov. 2016. 842K posts — of those, 800K in the last month (Hurricane Matthew). Can separate out “sub-memes,” map by geolocation.

One topic customer was interested in was drone usage.

Comparison with flood incidents in Austin, Tex., in May 2015. More influential posters on the Hampton Roads incidents than the Austin ones. But more emergency management posts on Austin.

Media, government, and citizens all participated in information sharing. Drone footage was useful. Twitter was used heavily. Power outages may send people out with cell phones to document conditions.


Discovery and Discoverability #InternetLibrarian

Marshall Breeding

Different models:

* World-scale – Everything in results (Libraries spend a lot of article databases, want them included in results. Not too expensive or difficult to index billions of items any more.)
* Bento-box
* E-book integration

Patron privacy concerns
Vital that libraries implement end-to-end encryption (https, TLS)
Third-party sites (such as Google Analytics): make sure user’s info is anonymized. Ghostery is a tool users can use to tell them how they are being tracked.

Most public libraries using online catalog or discovery module that came with their ILS, but some are using different ones.

Most public libraries use some kind of e-book lending. Want it to be integrated into catalog, not just a link.

Jarring to go from a library web site to a catalog that looks different.

Can assemble something with Drupal. The less you have to work with the code, the better.

Want everything under the library’s domain; most essential part of library’s brand.

Various systems use headings, keywords, concepts.

Discoverability: experiments with linked data.

(Google Structured Data Testing Tool)

Room for better, more intuitive display, linked data to supplement.

(ROI) Truth to Power: Measuring & Talking about What Matters #InternetLibrarian @mebs

Mary Ellen Bates

Slides here:

“What gets measured matters,” but
“Not everything that can be counted counts.”

How data vs. Why data
How data may be useful for internal staff, but it’s not ROI.

Outcome data: the value, the impact of what you’re doing.
(Referring to previous session: being a trusted advisor may take a lot of time, but if you’re doing it for someone high up in your organization, that’s the best ROI you can get.)

Cost of material vs. circulation:
20% of cost * # circulation = value (assuming they would have paid at least 20% for a used book)
Journal routing saves $ vs. article purchases
ILL also saves article purchases (c. $40)

Show impact on revenue:
Supporting your city council
Student success
Patron engagement goals
Solicit testimonials of impact (especially when you do a big-deal research project). You can just send an e-mail; people don’t usually volunteer their gratitude.

Show impact on your organization:

Supporting employee development -> improved employee retention. (A tech company had employees bringing kids to work, so the library included children’s books. People have cited the library as a reason they’ve stayed with the company.)

Effective outreach to stakeholders resulted in [an action on stakeholder’s part]. Why did they contact the library? It was in order that they could do something. What was that thing?

Show impact on org’s staff:

Look at information flows, pain points. Librarians are “information whisperers.”

* Time spent searching (and not finding)
* Duplication of effort within team
* Underutilization of resources

A librarian watches for search boxes on intranet and takes it upon herself to give advice.

True cost of your time:

* annual salary * 1.3 = fully loaded (with benefits, etc.)
* 52 weeks – 4 weeks = 1,920 work hours/year (vacation, holidays)

Salary / work hours = hourly rate

Many of the people we support get paid more than we do.

Outsell said a library interaction saved a user 9 hours.

If 1 hour of your time (@ $68) saves 9 hours of someone else’s time (@ $102), you saved $850.

If you teach people things like how to use your subscription databases vs. fruitless Googling, you saved a lot of money over the course of a year. Same thing if you created a UI to make it easier and more effective to use.

What are your org’s strategic goals for this year?
Do you read your org’s press releases? Do you follow their social media?
Do you follow new ad hoc teams that are set up to do things? (Suggest targeted services for them.)
Read between the lines.
For-profit: increasing no. Of products
University: student job-placement rates
Non-profit: Strengthening relationships with partners

Describing the less-measurable:
Ask users why they are asking that question. (Ask nicely.)
What will this be used for? What’s happening to this information next? (Might also help you format the information so it’s most useful. Help you put it into a presentation, inform a team, guide a decision.)

If your deliverable isn’t frictionless, your clients will go elsewhere.

Follow up after high-value research project:
* What difference did the library make for this project? (Even if the answer is bad news, you can ask what you could do better. And you don’t have to put it in your report!)
* What impact did this make for your outcome?
* What would it cost to achieve [your goal] without [our services]?

Look for programming that will help people.

Embed widgets at pain points
Embed content
Curated daily news!

Look for underutilized resources. Promote it to likely users.

Face-to-face contacts. Feed stuff to “ambassadors” of your services.

Use new language:
Describe yourself by outcome, not activity
“We’re here to make you more successful. What do you need?”

Not: We centralize acquisitions, but we save money.

Librarian as Consultant #InternetLibrarian

Paul Barrows, Federal Resrve Bank of San Francisco


Pyramid: transacter -> problem solver -> consultant -> trusted advisor

Problem solver: more involved than a simple reference question.

Consultant: someone keeps coming back with a subject.

Trusted advisor: someonce comes to you when they’re really lost and they trust you to get them started.

The pyramid narrows at the top, because you can’t be a trusted advisor to very many people.

New MLIS: “My boss is very smart, she must have had a reason for hiring me.”

Perpetual curiosity, ask lots of questions, be a team member, speak up, change leader.

Fewer ready reference questions, needed to transform services.

Positioning the librarians as advisors, partners, consultants.

Transforming staff: re-training, tough decisions, playing to strengths, honing existing skills, emerging skills.

Think about your mission as furthering the success of your parent organization, rather than self-preservation.

Find about your organization’s mission, your management’s priorities. Commitment to customers’ goals (they want to look good and do well), future of librarianship, and your own professional development.

Librarians are already generally trusted, but people don’t know what we do. People think they’re bothering you. Soft skills, like empathy and emotional intelligence (know the difference between panic and curiosity!). Get in on the ground floor of projects. Iterative approach improves the product — and the relationship (don’t give all at once, make sure you’re on the right track). Regular brief meetings during larger projects. Over-deliver and maybe offer more. Learn and ask about something personal.

Build awareness of services, esp. online subscriptions. Have a strong web site, self service but can come to us for deeper levels of service. Pop-up tables.

Catch key clients: Meet with new executives and give them targeted recommendations. Presentations to departments and divisions. People are happy to find out you can help them do their job better. Check-in regularly: “What are you working on now?”

Trust your gut, even in the deep end of the pool.
Periodic SWOT analyses (organizational and personal)
Are all your strengths being used?
Go on field trips to where people are
Polish how you talk about the library, yourself, and your colleagues
Influence without authority (peers, executives)
Suggest the wacky if you can explain how it serves the mission
(Something for everyone is part of their goals.)

Secret Sauce of Search #InternetLibrarian

Marydee Ojala, editor of Online Searcher

Presentation slides

Our work starts where Google ends.
Anybody can Google, but not everybody searches well.
Search does not equal Google.

Other search engines:
Country versions of Google
Duck Duck Go
Peekier (another one devoted to privacy)
Wolfram alpha
Million Short
Similar Sites
Wayback Machine (

Advanced search:
Special syntax, prefixes (site:, filetype:, inurl:) (Bing and Yandex have others)
Phrase searching
Word order, synonyms, language

Non-textual search:
Images, audio, video, datasets
Specific databases at Google, Bing, Yandex
* YouTube, vimeo
* Flickr, Morguefile
* Zanran, Datahub

Specialized search engines:

Topic specific:
* Biznar
* (market research)
* PubMed

Academic search engines:
* Google Scholar
* Microsoft Academic
* Semantic Scholar
* MetaBus

Academic document delivery:
* ResearchGate
* Sci-Hub (a pirate site)

Subscription search engines

* combating fakes and frauds
* look for additional documentation
* what is the source?
* Not every issue has two sides
* Not just fake news
* AllSides – News from left, center, and right political viewpoints

Being ethical:
* Doing the right thing
* Teaching people about copyright
* General Data Protection Regulation (GDPR) – protecting people’s personal info (European regulation, coming in 2018)

Going under the hood:
Knowing how search works
Search technologies
Machine learning, AI
Semantic search, contextual
Moving away from keywords
Why did I get this result?
Why are these ads following me around?

Google decides to disregard some of your search terms and puts a note telling you what’s missing.

Secret sauce:
Knowledge about search
Be willing to experiment
Think non-linearly, accept imprecesion
Constant updating of our brains
Power of the info pro

Updated to add link to slides.

Practitioner’s Panel: Search Tips and Millenial Searcher Secrets #InternetLibrarian @aainfopro

Amy Affelt
Slides at

Fake news. Shared knowing it’s fake, shared without reading page linked to.

Facebook said it was crazy to think it had anything to do with influencing U.S. election. Then said maybe it did. Has some remedies, but they are much too slow.

Google has automatic links to face-checking sites. But are those correct? Or consistent?

CNN pointed to an IFLA document for spotting fake news.

Look at domain names. Read “about us” pages.

But respectable sites aren’t always correct. CBS News said Tom Petty died before it was true. Look for supporting sources. One source may not be enough.

Read about the author.

Check the date. Is it old? Is it April 1?

Check your bias. Does the site have a political slant?

Watch out for provocative headlines.

Watch for the “promoted” label.

Fake health news: Check source, plausibility. “The secret doctors won’t tell you.” Look for peer-reviewed article, human trials. debunks fake health news. Google it for other reactions.

Melissa Zimdars’ list of dubious news sources.

How to Lie with Statistics. Book from 1950s showing misleading graphics.

Fake videos coming next.

Tom Reamy:


Text analytics, taxonomy, training.

Book, “Deep Text”

There is no single definition of “fake news.”

Fake people, automatic bots (e.g., on Twitter)

Google: can be manipulated so top stories on a topic are fake news.

Two drivers: making money and manipulating people.

People make up news to get clicks, which get ad revenue.

Debunking: a fraction of the people who saw the original post will see the debunking one. No money in it. Effects linger.

Can block ads on a site, but that does nothing for politically motivated fake news.

Technical tools to finding misleading domain names, etc.

Automated systems aren’t smart enough and can be manipulated.

Case study: hybrid analysis of news
Inxight Smart Discovery (now SAP), multiple taxonomies.

Pulled in thousands of news stories, used rules to categorize them.

Faster than human review, smarter than automatic solutions.

Weight if word occurs in title, etc. There is no such thing as unstructured text.

Pronoun analysis: “The Secret Life of Pronouns.” All the words that search engines throw out. Analyzed e-mail and could often tell age, gender, power status of writer. Lying and fraud detection: fewer and shorter words, more positive emotion words. C. 76% accuracy.

Deep text solutions:
1. Database of known sites.
2. Deep learning of text/linguistic patterns
3. Flexible categorization rules

Fake news is a serious problem: undermines democracy, communication, civilization.

Multiple factors: multiple solutions

Hybrid human-machine solutions are getting best results. Ultimate solution is better education.


“Weaponized lies”
“Don’t Think of an Elephant”

Libraries can keep pushing information about how to spot fake news.  Colleges and universities can teach critical thinking.

Take a minute to check it out.  You don’t want to be “first but wrong.”