Notes On One Page (2009)

This page contains all the student notes from the whole semester.


02 Web Search

by dabilendabilen (14 Sep 2009 16:45; last edited on 07 Dec 2009 15:53)

BIT 330 Class Notes – September 14, 2009

I. Signing Up for Assignments

  • Login to Wikidot, then return to www.howcanifindit.com
  • Direct your browser to the bottom of the assignment page and click “Edit”
  • Enter your username into the slot you would like in following format
    • [ [*user insert username] ]

II. Twitter Updates

twitter-logo.jpg
  • Get a twitter account
  • Watch out for assignment update/reminders on Twitter
  • Use Twitter as a resource to send questions to Prof. Scott Moore

III. Search Process

  • Future of Search: Tools for search are constantly being updated
    • Experts used to be the only wielders of information, “the old way”
    • Experts are now You & Me, the world of wikis supports this new era of information
  • Search Tools: search engines interacts with document(s)
    • Process: Query from User > Interacts with Document Set > Output Query Results > Possible Further Interaction between User and Engine
  • Search Engines Technology – almost all search engines are “Default And”
    • General rule, add “more” to end of query to get more broad results
    • Google Spider – Program to constantly search for new information as it is published online, constantly keeping Google on the cutting edge of information awareness online
    • Directory/Meta-Search Engines – Gathers information about websites, does not keep whole website on database
    • Main 3 Engines we will look at: Google, Bing, Ask
  • Search Terms and Operators
    • ~ [tilda]: searches for synonyms
    • “ ”: phrase search
    • –: terms to exclude
    • +: search exactly as is
  • Evaluating a Search Engine’s Performance
    • Query results will have an undefined return of both relevant and irrelevant results
    • All results of a search engine query are the retrieved results
  1. Now there will be some relevant documents that are not returned and some irrelevant articles that will be retrieved
  • Search Experience – How to judge the relevance of a search engine - Important for Exam!
    • Query – automation, variety & usefulness of special queries
    • Results – content, format of results, delivery form
    • Subset of Web – opacity, target, quality of coverage
    • Searchable Information – how frequently is the information updated?
    • Search Engine – quality of search experience, responsiveness

04 Search Technique Notes

by joshuaaajoshuaaa (21 Sep 2009 18:44; last edited on 09 Nov 2009 16:36)

by joshuaaajoshuaaa

General Questions

  • In blogs, the title can be whatever you want it, but the last part of the URL has to be the [blog: title]
  • Bloglist is a program that goes through all the blogs and orders them in a particular fashion
  • Easy to rename pages, down at bottom of page: click options, rename
  • It sometimes may be easier to use a text editor rather than type in your web pages

Don’t forget, bullets are asterisks and numbered lists are pound signs

How to Create a Blog Entry

  • Type "blog:name of blog" at the end of your personal wiki page URL (or type it in the “add new page” box)

Why Do Search Engines Return Different Results?

  • Different files that they index
  • Different search syntax accepted
  • Processing of search results differ, different back-ends

Why Use a Search Engine?

  • The main point of using a search engine is to find information fast
  • Goal of class is to become masters of search techniques
  • The simplest tools are the words and phrases you type in
  • Some operators target specific parts of a document (i.e. intitle:subject)
  • Others target the URL
    • flintstones inurl:umich OR inurl:lib
  • Some target sites
    • flintstones site:www.lib.umich.edu OR site:umich.edu

Helpful hints

  • Unique words and phrases are key
  • Make your queries as specific as possible, as early as possible
  • You have to realize the different ways things can be named or referred to
    • Formal references: last name, first name
    • Places can be more problematic
      • Use zip or area codes

Search Engine Practice

  • Use related searches to narrow your query
  • Open queries in new window – less easy to close a query
  • Try to think of what the document would be called or what kind of information it would contain when searching
  • Directory function of Yahoo is like “tree tagging” – not random, but in categories
  • [Define] function returns specific results, just like weather or map
  • Ideally, put what you would like to find in your results as your query

To view Dr. Moore's outline for this class, click here


05 More Search Techniques

by keyurpatkeyurpat (23 Sep 2009 15:34; last edited on 07 Oct 2009 13:55)

No Formal Presentation Today

"How many" Queries

  • While using quotations in queries, each search engine we have explored has the limitation of only being able find exact phrases without much flexibility
  • A useful tool for creating your queries is the * symbol which allows the search engine to insert a word or short phrase within the query

Example #1

  • Let's start by looking at some of the exercises from today's class:
    • Come up with a list of 30-40 cities in Michigan.
      • Start with this query
      • Looking at the results, we get a bunch of pages describing the number of cities in Michigan that . Scrolling to the 5th link down, we see an exact match. Looks like we lucked out this time.

HINT: More often than not, a query will not yield what you are looking for on your first try. You should use these exercises to practice your ability to form good ones.

Example #2

  • Let's look at another example:
    • How many people are in the United States?
      • We'll start with a basic query
      • Looking at the results, we can conclude that a number of pages cite the U.S. Census Bureau as their source of information. This could be useful! However, one question that arises is whether the numbers presented are up-to-date. Let's try putting both of these ideas together by looking for a 2009 census or at the very least a number from 2009.
      • Let's apply our newfound knowledge. Try this query
      • What did we get? Scrolling through the page, we see more relevant results. However, we are still seeing a bunch of irrelevant numbers not pertaining to the entire U.S. population. Some numbers aren't even from 2009. But how can that be? Looking at the urls from the last query, we can see that searching for 2009 within the urls could fix part of the problem. Let's try using this information.
      • Here's a new query
      • Look at that! We got a handful of different sources all reporting a similar number…and they're all from 2009! Clicking on "repeat search with omitted results" will yield 14 different sources. Most importantly, we have arrived at the answer we were looking for.

HINT: When developing your query, try to assemble a handful of key phrases to incorporate into your search

Summary

  • As you can see, by combining a number of different tools, we were able to arrive at many of the exact solutions we were looking for.
  • Remember that your tools are not limited to simply what's put in the query. Look at the relevant results: are there common urls, titles, terms, etc.? You should be thinking about all of these things in order to incorporate this information into your next query.
  • If the information from your relevant sources is not sufficient, take a look at the irrelevant results. What do they have common? Incorporate this into your new query.
  • Rinse, lather and repeat. Sometimes it's hard to get that query right on the first try, but keep trying! With practice, you'll be able to become much more efficient.

Homework

Remember to do the experiment and submit it by 5pm on Sunday!

Reference

To view Dr. Moore's outline for this class, click here


06 RSS Introduction

by ranfanranfan (28 Sep 2009 03:19; last edited on 07 Oct 2009 13:52)

This class introduced the topic of RSS feeds, blogs, subscribing to RSS feeds, and the basics of searching for RSS feeds.

Updates to Course Wiki

  • Announcement Page now has all the recent Tweets through Tinker
  • Wiki will sometimes be down, plan accordingly!

Lecture: RSS Basics

  • Preview: RSS will change your life: It'll save tons of time by pulling info you're interested in all to one location and delivering it to you
    • Rapid growth of info + Lack of time + High demand for info = Headache!

RSS does the work for you! Helps you find information and sends it to you. YAY

What is RSS?

  • HTML Document: bunch of tags (e.g. <p> = paragraph), looks messy and unstructured
  • RSS is a text-based, XML-based standard (meta standard)
    • RSS is also a bunch of "tags", but it is more structured in its format (it is standard)
      • <item>, <title>, <link>, </description>, </description>, <pub0note>, </item>
    • The tags above are set / agreed upon. These tags are consistent across all RSS feeds.
  • RSS is an open, relatively mature standard developed fo rhte purpose of tracking site updates
    • Different standards out there for RSS Feeds:RSS 0.9, RSS 1., RSS 2., ATOM; pretty similar to each other

How do RSS Aggregators Work?

  1. Lots of servers connected to internet
  2. You are on a browser with a RSS account
  3. RSS aggregator has your RSS acct name and list of RSS feeds you are subscribed in
  4. Lots of docs in those servers that you are interested in reading
  5. Life before RSS: go (how ever frequently) to those websites to read what's new
  6. RSS aggregator gathers all those docs FOR you and delivered to one website to which you go to read updates

RSS Standard: having a common standard means lots of RSS Tools can be specially written to work with it. Site development Tools means all Web Sites can use RSS meaning you can get the info via RSS aggregators

Web based feed readers are needed to get you started

  • RSS reader focused tools (e.g. Blogline, Google Reader)
  • Portal type sites: bring info to site besides just RSS feeds (e.g. iGoogle)
  • Others

Searchable subject indices of RSS feeds (with browsing) helps you discover the most popular feeds

  • Searchable feed databases help you find the hidden gems that you need (e.g. Technorati, Blogpulse).

RSS tools give you a way to multiply your own efforts, save time, manage your attention, improves ability to find info,

Get started with Google Reader or bloglines.

Class Exercises

Searching for blogs using bloglines

  • I searched for the term [U.S. healthcare], and it returned 270,500 posts with Matching Feeds and Matching News to the right, with option to sort by relevance, date, or popularity
  • The different search options (posts, feeds, citations, the Web, or a specific URL) for when I have an idea of what I might want

Searching for blogs using several different blog search engines

I compared different tools to see what kind of information would be returned.

  • Google Blog Search
    • Offers "Related blogs", "Sorted by relevance" vs "Sorted by date" (It also has a published within the last xx amount of time period filter option to the left and a "browse top stories")
    • If I want more information or want to follow up, I can use:
      • Email alert
      • Blog search gadget
      • Blog search feed for Google Reader
      • Search Google News
  • Technorati
    • Can edit "authority" for searching (who determines authority?)
  • Blogpulse
    • Search for [iphone android] Trend
    • Can see time of increased activity of discussion and overall growth trend
20090928123800FsDjXKNxcabXywcNm7El.png
  • Blogs.com
    • Search for [google android apple iphone].
      • The top few posts are in reverse chronological order (from most recent to earlier) although the "earliest" starts in August. There had to have been some discussion on the iphone within the last month!
  • IceRocket
    • Search for [obama healthcare]
    • "Results Trend" shows a date vs percentage of all blog post graph between 8/30 and 9/2 - It peaked around 9/7, 9/8
      • There were 450 posts a day, On average 0.077%, and total of 13,501 posts

See Professor's Notes

See Resources

  • Online feed readers, Top lists, RSS feed databases, searchable indices, keyword based feeds) available towards the bottom of the class page

Action Items:

  • Talk to professor about term project topic during OH! (MTW, 3:30-4:30)
  • Update the student list page after talking to professor
  • Add yourself to the Grade Database if you haven't already
  • Make Shared Items page on your wiki concerning term project

08 News Search

by WdchenWdchen (05 Oct 2009 15:47; last edited on 23 Nov 2009 15:04)

Introduction

  • There was a reorganization of the Course Website so be sure to check it out. It now includes a Current To-do List so hopefully that helps with trying to figure out when assignments are due and reduces the amount of confusion.

Today's Class

  • Today, as Professor Moore states, is best day for class exercises because it provides tools that can be used by literally ANYONE
  • The main focus of today's class is to go over the various news search engines and how they can be applicable to not only your daily life but also to the searches for our term project.
  • Listed below are some of the various News Providers, but of course the list is not limited to only these. Please feel free to add anymore as you find them!

The more you know, the better off you are. Knowledge can never hurt you, so keep updated!

Searching for News

  • Using one of the tools stated above will help you to figure out a way to stay on top of the news. All of these sites do all of the work for you. You only have to enter in a search query!
  • There are numerous resources out there that will put together a list of news articles that you are interested in. Utilize them properly to stay current and up-to-date with the events that are happening all around you.
  • Why is a News Search different?
    • It allows you to find legitimate sources for news.
    • The articles provide with you the following information, such as, who wrote it and what it is written for.
  • 4 Desired Features for News Searches
    1. Response time
        • Before technology became so advanced, it would be faster to wait for the evening news than to go online and search for the newest and latest breaking news
    2. Relevance
      • After submitting a query, the results for a news search tend to be pretty accurate and relevant to what you are looking for. Of course, it might depend on the topic, but for the most part, news searches will generate quality results.
    3. Update speed
      • Essentially, we are able to find out the latest news as it happens in real time. Within minutes of the event, articles about it will quickly be posted on these various search sites.
    4. Quality of sources
      • If you're getting news articles, you want them from a voice you can trust.
  • thief.gif

News Searches

  • Word Phrases
    • In order to generate more accurate results, you can expand the scope by paying attention to these key words.
  • Date
    • All of the news articles are sorted by the date the news article was published, so you don't have to scroll around to find the most recent.
  • Event
    • In terms of a news story, we often think about the following: Who, What When, Where, and Why? These provide you with accurate and pertinent information regarding the event.
  • Source
    • The source is the article's origin. Where is this article coming from? Is it from a credible source?
  • Source location
    • This is essentially the location from where they are featuring the story from.

Special Features

  • Ranking/Sorting
  • About location
  • RSS Feeds
    • With the RSS Feeds, you will be able to subscribe to the news that interests or is relevant to you. Therefore, you will be able to have a steady stream of articles directly sent to your feeder, without you have to do any type of query.
  • Categories
    • Search on stories to help weed out irrelevant results
  • Author
    • Helps to find stories, such as, by: Tom Friedman instead of about Tom Friedman

Tools

News Sites
  • Google News
    • This application features news articles that are constantly updated so that readers are able to browse through and find the latest news
    • Features various sections depending on what kind of news you want, such as World, Entertainment, Sport, etc..
  • Yahoo News
    • Has a similar set up to Google News providing you with the latest news and then various categories for you do to a quick scan.
  • Google Fast Flip
    • Similar to a PDF representation of the "first page" of website
    • Fast Flip allows you to do a quick scan of the news articles and if one particularly interests you, then just click it to view.
  • Google News Timeline
    • These news articles have been scanned and archived from past years.
    • They allow you to go back in time and read up on past events.
Analysis Sites
  • Silobreaker
    • Searches are based on using keywords
    • The results are featured in a Textual analysis, which help you to understand what the story is about
  • Newssift
    • Searches are based on relevant topics.
    • The results are listed in a categorical manner, where you are able to do a quick glance and find articles that you are interested in or will be able to use for the term project.

Class Exercises

These exercises will help you utilize and get more efficient with these news searches. It would be a good idea to work on these to help you get used to these various tools.

Class Notes

The reason that I have these posted is just so you are able to reference back to everything we did today. Hopefully the instructions have been clear enough for you to get a good idea of how to make the most of News Searches

Actions Items

We have learned most of the tools that we will be needing for our term project, so keep on plugging away at collecting information. We will need to be using relevant data throughout the span of this semester, so you can't procrastinate! Continue saving the relevant searches and collecting data that will be useful for the project. These can be saved either on your chose RSS Reader, or for single articles, since we haven't learned to filter yet, you can save them on your Notes section in your Wikidot.


12 Research Sites

by tfornerotfornero (21 Oct 2009 15:48; last edited on 21 Oct 2009 16:17)

Class 12- Research Sites

Focus on Academic research sites, basically sites to help you cite.

So far we have been focusing on the surface web, we've been connected to a bunch of html pages (they're available right there at the "Surface"). Today we will talk about the "hidden web" or the "deep web". The idea is that there are query forms and behind that there are documents.

For Example: You go to the library and search by a query form (Title, Author) so instead of a page for each book, they have a form that gets you to each books. 60 largest deep website contain approximatley 750 terrabytes of data (BIG) "Is it enough to care about… YES" Estimated to be about 300,000 query databases out there.

Why do we care?? Does Google Go to it? NOPE. So we are ignoring vast subsets of data that we could use. So we need to use specialized search engines.

Specialized search engines have three important attributes 3

1. Comprehensive - When you go to a deep web search engine that it gets enough of (you'd love all) journals or websites you would like it to reach.
2. Integrated - Think about all the different kinds of forms out there they you can fill out and get data from, you'd like to be able to type something into some form and the search engine can use that to fill out all the different variants of forms (difficult). Integrated to work with all different forms equally well.
3. Transparent - When you go to the deep web search engine you want to know what you are search engine is searching (what journals its searching etc.) Google Scholar doesn't tell you

Looking at 4 Different Kind of Sites

1. Deep Web Search tools
a. General
b. Specialized (ex. Science, business, medicine)
2. Library and Book Based Sites
3. Web Directories - a evaluated and vetted set of resources that are categorized (someone has already deemed sites to be useful)
4. General References and Answers

Examples:

Google Scholar

  • When first came out five years everyone thought it would be a great academic search tool
  • Looks different (Articles, Authors, Were it comes from, Cited by)
  • You can setup so it says when these documents are available at your library
  • Can limit searches to certain date ranges and just get summaries or citations
  • Used to be some real problems but it's been fixed
    • Example: Search for A = x Search for The = Y Search for A or the = only x (Fixed though)
    • Last years theme was that Google Scholar was the worst piece of software ever, fingers were threatened to be broken. A day's worth of work has made big changes
    • Can search intitle: or allintitle: but also can search only within subject areas (Can't do in Google itself)
      • Some documents aren't categorized and will lose them if you specify but subjects (even if you select all subjects!)
  • Can specify certain dates
  • Use it as a nice place to start, but be careful

Scirus

  • Nice tool "Works like a champ"
  • Doesn't cover as much as Google Scholar
  • Limit to certain Journal Sources, Date, Refinement

Google Books

  • If anyone ever did a report on Google Books would keep you busy on a day by day basis
  • Huge story because Google thought it was right and good to take out of copyright and scanning and put on web (easier access). Some people saw this as infringing on authors rights and evil (Contradictory to Googles "Do no Evil" policy)
  • Searches give you results about whole books, reports, and authors

Search Tool at U of M Library

  • Will in real time search through multiple databases (can also use tags to get into data)

Biznar

  • Business deep web site (specialized)
  • Information should be pretty reliable (and filterable)
  • Same thing at Bnet

WolframAlpha

  • Not a search engine, but it attempts to answer the question it thinks your are trying to ask

Yahoo Directory

  • Can go into Business Economy
    • Example you search Airline Industry and it will return to you directory categories

13 Change Notification Tools

by oish330oish330 (26 Oct 2009 16:40; last edited on 28 Oct 2009 01:44)

Monitoring Changes

There are millions (Google says ~ 206 million) of sites that provide email alerts. Unlike RSS feeds, these come through your email inbox, but similarly you can go out and find what types of alerts you would be interested in and receive updates about those topics. The difference is that these are typically only sent out daily, weekly, or by some other constant time interval

Site Based vs. Search Based

  • Site Based = already set up alerts for you
  • Search Based = you set up the alerts based on a search of interest

There are also services like Google Alerts and Yahoo Alerts that will only send alerts via email when something is updated.

  • Yahoo! Alerts has already setup topics that you can subscribe to.
  • Google Alerts only allows you to set up email alerts for search topics of your own (nothing already set up).
  • The nice thing about using these is that you can customize a lot of the details, such as how often you want to receive updates.

Page Monitoring

Page monitoring can either be software-based or web-based. We will cover web-based methods.

Why would you want to page monitor?

  • EX: interested in Goldman Sachs press releases? They do not have an RSS feed or email alert system, so you could set up a page monitor to alert you whenever there is a press release. This can save you time from checking to find out when new things are added.
  • Other ideas: find out when new TVs come out on Amazon, track movie comments, or follow job listings of a specific company.
  • One other cool idea is to use page monitoring to track who is following your own website (or a website you follow). Just search for sites linking the site of interest and then set up a page monitor on that search.
    • Say that you want to track who is linking David Pogue’s blog:
      • Search on Google for the query: link:pogue.blogs.nytimes.com/ -site:pogue.blogs.nytimes.com
      • Set up your page monitor to track that search.

What to use for page monitoring?

  • WatchThatPage.com: Free (for any number of pages), or $20/year for priority service
  • TrackEngine.com: Free for 5 bookmarks, or $20/year for 10 pages, or $59/year for 50 pages

Feed Creation Software

Can be used to filter information and then send it through an RSS reader. Let’s say you’re interested in Google, Amazon, and Apple. You can create a custom feed that filters the NYTimes “companies” RSS feed for news on just these three companies. Try Feedrinse to do this.

Yahoo Pipes
A visual method to combine different feeds, filter them, and then output to an RSS feed. The information that comes out of the “pipe” will then be displayed on your page.

  • You can also search for already created pipes that may be of interest.
    • If you happen to find one someone else created that you like, clone it and then edit for a customized version.
  • View options allow you to see image or text-based.

Google Social Search
If you have a bunch of RSS feeds that don’t send a lot of information through frequently, you can use this to compile them so that you won’t need 15 feeds to get specific information.

Convert to RSS

Take structured web pages and use one of the tools below to convert to an RSS feed.
Dapper – Dapper is probably the coolest of the three. We will look at it a little bit more in depth. Make sure not to go to their homepage, but rather dapper.net/open

  1. Collect sample pages
  2. Dapper looks at these pages and tries to find commonalities in their structure
  3. Select Content – You can pick titles, description text, and URL, but remove other parts that you do not want to see in your feed.
  4. Dapper will create an RSS feed based on your content selections.
  5. Review and save to your profile.

After creating the feed, you can update the input with new search terms and Dapper will create an RSS feed based on this search using the structure you just set up.
Feed43 – A bit more complex than Dapper. It will see text patterns.
Feed Yes

Email Filtering

“A hint that works sometimes.” Can be used to filter just some information and do different things to it (for example, in Gmail you can forward only some of the mail to your phone or have Gmail tag different types of emails).

Lastly, you can check out the class 13 page for ideas on how you can use the information we talked about in class to help with your term project, some possible blog topics, and hints about possible test questions.

Action Items

  1. Wednesday 10/27 is the due date for the demonstration portion of our Search Engine Analyses.
  2. Keep working on your term project, trying to tie in the information we have been learning.

14 Custom Search Engine

by nikguptanikgupta (28 Oct 2009 15:56; last edited on 28 Oct 2009 16:52)

Note: additions to his notes are in bold.

Class Discussion

No notes here. Extensive discussion about how class is going, what should change, etc.

My class notes

Today we're doing custom search engines. You can basically get your own search engine that you control.
Google Custom Search Engines, which essentially took over the industry, changed in a huge way this Monday. Everything Prof. Moore knew about it, changed.
Prof. Moore's Search Engine:

  • Gives list of included sites.
  • His search engine keywords: nutrition, health, diet, food

What is it

  • A search tool that uses Google's search engine (as the back end) but that you can instruct the following ways:
    • Look in a certain list of URLs (pages, whole sites, or subsets of sites) (what to look at)
    • Avoid a certain list of URLs (what not to look at)
    • Append a set of terms to any user-supplied query
    • Customize its looks (within bounds)

We're going to create a custom search engine for your project today.
Then, we'll learn how to embed the search engine in your wiki. Lots of different ways because each of them will work under dif. circumstances and with different tools, i.e. blogger, wordpress, wikidot

How can it be used (3 different ways)

  • Its own Web page
  • An iGoogle widget
  • Embedded in random Web page
      • See today's exercises for instructions on how to embed
      • Google Custom Search Engine is an easy way to do it.
      • Note: IFrame doesn't work on Firefox for Mac and Windows. Skip it.
      • Search Element Based Process- looks the best, provides the best user interface
        • Three ways: full width, compact, and two-columns (trouble implementing it)

Why would you use it

  • Captures creator's knowledge of the field

In previous years students have explored and learned about several search engines: Topicle, Eurekster Swicki, and RollYO; however, for the last year or so these sites have been completely displaced and dominated by the Google Custom Search Engine so that is the only tool we are going to look at today.

Blog topics

  1. Describe how useful or not Google Custom Search Engine is for your site.
  2. Describe how you chose the sites to include in your custom search engine.
  3. Compare and contrast 2 (or more) different custom search engines.

Resources

  1. Topicle
  2. Eurekster Swicki
  3. Rollyo
  4. Google Custom Search Engine
  5. BuildASearch
  6. Yahoo Search BOSS — Build your Own Search Service
  7. 3 Guides to FireFox Quick Searches (Smart Keywords)

15 Project Day

by npajanknpajank (02 Nov 2009 17:55; last edited on 02 Nov 2009 17:55)

Creating a list that updates automatically

To create a list that updates automatically such as the bloglist on the homepage of www.howcanifindit.com, you can do the following:

  • Click on "How to edit pages" on the left menu
  • Click on "Documentation pages"
  • Click on "Modules'
  • Click on ListPages under Listing and categorizing pages
  • Go to "Ordering Pages"

There you will find different ways to list pages using a module

If your page disappears

  • Try putting "/norender/true" after the URL
  • If that doesn't work, twitter Professor Moore

16 Wikidot Day

by jenmmartjenmmart (04 Nov 2009 17:15; last edited on 04 Nov 2009 17:50)

Website Popularity

  1. Are you wondering if the website you're using is well known, oft-used, or dying out?
    1. You can go into Google Trends to look for websites (hmm, digg.com is dying)
      1. If you sign in, you can actually get the volume scales on the side of the graphs, telling you the daily unique visitors
    2. You can also use Alexa
      1. It can show you for yesterday, 7-day, 1-month, etc.
      2. Options include percent of global users who visit the site, traffic rank, page views, etc.

Wikidot Issue and Options

  1. Can we make templates for things besides "blog:"?
    1. Change "blog:" to something else such as "news:" or "newsstories:"
  2. If you want to see how to do tables, look up coding for announcements under his homepage
  3. Images not showing up
    1. Make sure the image is all one word
    2. Particularly important if you're uploading files. For example, it should be "mypic", not "my pic"
  4. Header Images
    1. Notice his page has an image on top. We want images, too
      1. However, there will not be a grading box for the project that says "is beautiful and glorious", so don't worry
    2. There are lots of links on his page for this day
    3. The header code is what makes it possible to change the image on the top of the page
      1. Go to: http://www.wikidot.com/common--theme/**MYTHEMEHERE**/css/style.css
        1. This is the code for the theme you want to change
        2. The naming for the code makes it pretty easy to guess what part of the code it includes
      2. Go to: Site Manager -> Appearance -> Custom Themes
        1. You can create a new theme based off of a theme that you like (like "Bloo" or "Cappuccino")
        2. Use HTML Color Code to pick pretty colors for your page
        3. Make obvious changes so you know that it's working before you go changing little things
        4. It might not show up right away, you might have to reload the page, close your browser, or even turn off your computer
    4. Color Blender
      1. Gives you options of good color combos
    5. Wikidot Themes
      1. You have other options for themes, tons of them, most of them ugly (so he says)
    6. Wikidot Snippets
      1. Includes lots of tutorials on how to code certain things, like having an extra side bar, or lists, or layouts with tables, or sitemap, etc.
    7. Modules
      1. List pages, page calendars, etc.
    8. Templates
      1. Tons of different choices for making a template

That was a "bunch of junk", you can chose to ignore it and that's fine. Or you can make your wiki page look totally cool. Ball's in your court.


17 Image Search notes

by moonparkmoonpark (09 Nov 2009 17:20; last edited on 23 Nov 2009 14:55)

Introduction

11/9/2009
In class, we have outlined the numerous different ways to query images. The more traditional search tools include Google Images, Yahoo Images, and Ask Images, which tend to do a good job in locating desired images, but lack in search options for the users. However, there are varieties of alternate image search engines such as Flickr. Today’s lecture was based on these alternate image search tools: Google News Images, Yahoo News Images, Flickr, Compfight, Behold, Pixolu…

Types of Image Search Tools

  • Normal web query engines
  • News image
  • Obtaining similar image types
  • People face query
  • Up-to-date images

Please Note There are some search tools that you must subscribe/pay in order to use

Traditional Search Tools

Google Images:

  • Users can search the sites by size, type of image, description
  • Limited options/features for the user
  • Does a good job finding the necessary pictures
  • There is an advanced section for some other features

Yahoo Images

  • Filter by size, domain, and color
  • Similar to Google Image
  • Very limited in terms of user customization for search
  • Gives some side options during the query

Ask Images

  • Similar to Google Image and Yahoo Image

PicSearch

  • Can search for animations
  • Similar to Google Image and Yahoo Image

These are the tools that people use the most due to their high brand awareness and ease of use. However, these search engines might not be the best system to use depending on the situation. The list below goes into detail of very specialized image search engines.

News Image Sites

Yahoo News & Google News

These tools are used to find images regarding current events. Although they have tons of pictures in their archive, users cannot directly search for images. As a result, the searcher needs to type in the keywords and then click enter. It will then take the user to the result page where the top line will be filled with pictures. If the searcher wants to see more photos, there is a link right below the first picture, which will take the user to archive page. This site tends to be useful when the user is looking for news related pictures.

Flickr Tools

This is an alternate image search tool than the traditional ones mentioned above. The users can query by using full text or inputting keywords. In addition, the website provides tags for each of the pictures, which helps to locate pictures easily. There are tons of different features on this website, and should definitely advise to check out the tour section

Compfight

This is the Flickr search tool section. The user can search either based on text or tags, which are located on the left side of the search tool box. The results tend to be different for each of these features. If you are looking for a very specific item/object, users are advised to use the tagging feature rather than the text, which only covers the keyword section.

Behold

This is a useful tool that allows the users to search for files with a specific tag while narrowing the picture to certain criteria. As a result, the retrieved pictures tend to be very relevant to users’ needs. In addition, this search can locate images based on licenses.

Similar Images

Pixolu

This search engine is supposed to narrow users search query by refining the results based on users’ request. This is a powerful tool that scans through its archives to find similar images by elimination. The end results tend to be full of relevant pictures. Although there are other similar search tools such as GazoPa and TinEye, Pixolu seems to be the best at what it does. Furthermore, it has a large archive with many high quality images.

Face Search Tool

These tools specialize in locating photos/images of portraits.

Exalead

The results tend to be based on relevance and the lower the users scroll the less relevant the portraits will be. In addition, the site will continuously update the images as users scroll lower, which is useful since the surfer does not have to keep on clicking next. Other similar tools include Picitup and FaceSaerch, which are not as good as Exalead.

Real Time Image

PicFog, TwitCaps, and PicBrk

The pictures keep on updating on the screen based on the most recent post. These sites are not good at picking out the most relevant pictures nor are they good for good quality photos. Therefore, these are more for fun purposes rather than for an objective search.


18 Geography Based Sites notes

by mike_danhofmike_danhof (11 Nov 2009 17:07; last edited on 23 Nov 2009 14:55)

BEST CLASS INFO EVER!

  • New class to teach sophomores about business, what CEO's do, etc. We should not make the program better (haha j/k)
    • If successful, make required for all BBA students
  • Today is less about this class, and more about learning tools for other classes and your job
  • Geographic based sites have grown as mobile computing has become more and more widespread
    • Used to be engines that searched for pages from certain countries (Yahoo International)
  • Google Maps: changes all the time (Look at Google Maps Mania blog-really good and addicting)
  • Google Earth (favorite site EVER for Scott)
        • Sad story-Scott's pool at his childhood home broke, contractor went broke, never got to use (tear)
      • Look for Google Earth info whenever you are traveling

Travel Sites

  • Expedia, travelocity, orbitz are big. Kayak is big, been sponsoring some football game
  • The big ones have gotten smaller over the past few years. Kayak almost as busy as travelocity and orbitz
  • Local search has gotten huge in last 4 years; Ask City, MSN City Guides
      • Outside.In, yelp good for restaurant reviews

Road Trips

  • RoadsideAmerica maps shows cool/weird things you can see
  • Waze-twitter like updates about traffic

Public Transit-Google maps is getting really good, hopstop, NYC subway
Maps-just ridiculous amount of maps and progress over the years

  • UT has HUGE number of maps (historical, country, anything you would ever want)
  • Use CIA World Factbook (if you don't start here, you are wasting your time and being an idiot)
      • Google Moon/Mars maps (for those that will be billionaires someday)
  • Map Maker-build information about specific things (size of farms, roads, and more)—Play with this

Info Sites-use these for everyday information, research, and more

  • Auction mapper-good for local auctions (don't want piano from Cali)

Real Estate-getting amazingly good

  • Zillow-follow house value

Walking maps, mapping run, world times

  • All sorts of maps for all different kinds of activities, information, and just pure fun out there

And more and more and more and more…..


19 Video Search notes

by moonparkmoonpark (17 Nov 2009 00:10; last edited on 23 Nov 2009 14:55)

Introduction

Today’s class consisted of going over different search engines that specializes in videos. The video search tool websites have significantly changed over this past year (existing sites got better in terms of their features and experience). In addition, there has been a significant increase in the number of video search engines including Clicker.

Recent trends in this industry

The most abundant video search tool has been Google sites according to Score Video Metrix. There have recently been different approaches to how these search engines locate the relevant videos: web, deep-video search, and directories. The web based search is where the system looks for keyword or tags. On the other hand, the deep-video search actually breaks-down the video and analyzes its contents and audios for matches (much more fine grain). Lastly, the directories are where the videos are categorized into certain pool of groups. All of these contents can be applied both the web and site basis.

These categories can further be broken down to what the sites focus on such as entertainment. Furthermore, the way in which contents are shared can be different. Some sites allow users to upload their own videos while others allow to host, tag, or comment on videos. (Please look at Mr. Moore’s video search table for details)

Video Search Engines

Bing Video: They have just implemented a new feature where they broadcast news over the site.
Google Video: Added some additional options which can be seen on the left hand side after the search has been executed. There is an advanced search feature
Yahoo Video: There is an advance search feature and is based on keyword/description search.
Blinkx: Can run multiple video loads while they are embedded on the side bar of the screen. Pretty unique experience
Truveo: Aggregates different results from other websites and post them on the results page. There are many different options for this site such as narrowing the results by the source. They also categorize the video by ranks, most recent post, and popularity.
Pod-o-matic: Allows users to search by episodes or podcast. Users can also upload their own podcasts.

The other sites that people should be aware of are…
VideoSurf:
YouTube:
Daily Motion:
MegaVideo:
Metacafe:
Veoh:
Hulu:
Clicker:

Other

Please Note: There is a section on Podcasts, which we did not cover in class. However, there is a long summary of it on Mr. Moore’s note page. Please refer to this section for more details regarding Podcasts.

There are many other video search tools that you should be aware of including…

  • Yubby
  • YoVisto
  • CastTV

Furthermore, there are other search tools for music videos such as YouTube, Yahoo Audio. In addition, there are music search providers such as iTunes and Rhapsody


20 Metasearch

by rolay117rolay117 (18 Nov 2009 17:24; last edited on 02 Dec 2009 07:11)

Topic

Today's class covers a special type of search engine known as a metasearch engine. This is a search tool that combines the results from many different sites (Google, Bing, Yahoo!) in order to return the highest-quality results. Below are the two main types of metasearch engines

  1. Integrated results: This type of metasearch engine gathers results from the different search engines around the web and meshes them into one set of results. It is one search activity, accompanied by one set of search results. (ex. info.com)
  2. Unified-search engine: This type of metasearch engine allows you to search different topics within search tools (such as blogs, news, images) and then returns results from all of the websites. The difference is that the results are displayed separately by site, with sites separated by tabs on the results page. This lets you scan the different web site results without having to visit each site individually.

There is also another type of metasearch engine that is more of a hybrid type engine. This returns the results both separated by tabs and combined, whichever the user prefers.

Exercise

Today's class exercise involves visiting the various metasearch sites and comparing the search results to those of Google. This exercise aims to show use the benefit (or disadvantage) of using the metasearch sites. The exercise guidelines and results can be found here.

In addition to the exercise, we were assigned to post two twitters about our favorite two metasearch engines. Below is a list of some of the major metasearch tools and a brief description of what they do:

Tool Description
symbaloo Symbaloo is a type of metasearch tool. It's a page meant to be set as a homepage for your internet browser that gives you access to the many different search and media tools around the internet. Very cool tool.
search!o Search!O is a tool used to return results from many different sites separated by tabs. It's almost like tab browsing search sites, but within a web page. This allows you to browse the results of many different sites without having to actually visit each individual site.
soovle Fun, interactive search site. Useful when wanting to search a bunch of different, unique sources for one topic. For instance, finding youtube videos on a topic as well as visiting the wikipedia page for further information.
scour Nicely designed, easy to use metasearch tool. Nothing too special or unique about it, just returns accurate and easy to navigate results.
dogpile One of the most widely known metasearch tools. Searches for results on the main four web browsers (ask, yahoo!, Bing, Google)

To view a complete list of the metasearch tools, visit this site here.


21 Social Sites

by jadwinjadwin (23 Nov 2009 16:52; last edited on 29 Nov 2009 03:43)

Announcements

Class Notes - Social Sites

The benefit of social sites comes directly from people sharing.

  • Describing with tags
    • Taxonomies -
    • Folksonomies - classification derived from the practice and method of collaboratively creating and managing tags
  • Types of social sites
                      • Social News (based on tagging)
                        • Technology
                        • Search & Internet marketing
                        • For researchers & scientists
                      • Social Bookmarking (based on voting)
                      • Social activity (e.g., shopping)
  • Features & dimensions
                      • Voting - people vote on whether the site is good or not
                      • Current (e.g., "what's hot")
                      • Time periods (e.g., "last hour", "today", "last week", "last month") - find info not just right now but from the past
                      • Topic categories (e.g., "business", "entertainment")
                      • Tags (both personal and shared) - assign mulitple tags to these sites

General market size information

Delicious changed name, thinking they'd earn more market share, but as seen in the image from today's slides this didn't work. They are still doing better than their competitors though.

Resources

Social News sites

  • Digg is the big monster in this field. "Digg is a place for people to discover and share content across the web, from the biggest online destinations to the most obscure blog. Digg surfaces the best stuff as voted on by our users."
  • Yahoo! Buzz - Buzz articles up or down. See what people are buzzing about.
  • Reddit
  • StumbleUpon - A lot like Reddit
  • Fark: Basically just fun. They make fun of news reports.
  • Mixx - same thing as above
  • NewsVine - save thing as above
  • Slashdot "News for nerds. Stuff that matters."

Social Bookmarking sites

  • Delicious - Social Bookmarking. Allows you to search for titles, descriptions or tags and then bookmark the results you like. For just tags do "tag:yoursearch"
  • Diigo - "is a powerful research tool and a knowledge-sharing community." But it tends to be a little touchy. Diigo was foremost a powerful online research tool for highlighting, archiving, organizing, searching and accessing an article you found on the web. Really cool tool; can read article and highlight paragraphs of interests; add sticky notes on the page; and highlight pictures. In addition you can add tags to make things easier to find. Definitely watch the Tour Video.

Social activity site


22 People Search

by mkeagymkeagy (25 Nov 2009 20:18; last edited on 25 Nov 2009 20:34)

Today's class was all about how to find people on the web.

There are many different types of people search tools out there, making it easier to find:

  • White Pages Information
  • Criminal Information
  • Professional Information
  • Social Information (Facebook, MySpace, etc.)

Announcements

  • Happy Thanksgiving, eat lots of turkey.
  • Search analysis assignment is due today – make sure your link are correct on the sign-up page
    • You do not need to add anything to the site-maker page for this assignment
  • Exam is on December 2nd. Check out the Test page to find out information about it.
  • This Saturday at midnight (November 28th) is the last day to make modifications (change questions, notes, etc.)

Beginning of Class

  • Discuss the metasearch results from two classes ago.
    • Results were similar to last year
    • info.com was the only site to beat Google (two years in a row)
    • Clusty is crap
    • search.IO and Kosmix were the two favorites

People Search

  1. This problem has applicability to many areas
    • Just trying to find a person's address or phone number
    • Background checks (including criminal checks)
    • Finding missing persons
    • Ancestry (including obituary searches)
  2. A very frustrating, confusing topic
  3. Monetary interests have really infiltrated this set of tools
    • Lots of Web site purchases (industry consolidation)
    • Lots of pay-for-results

Tip: Some search engines get confused with middle initials

Exercises

  • Explore the 3 groups of people search sites
    • General people searches
    • White Page Directories
    • Social searches
  • Send 6 Tweets! Include your favorite and least favorite in each category.

General people search

  1. WhoZat (review)
  2. Pipl (about) — can search by name, email, username (screenname), and phone; emphasizes that it searches the Deep Web.
  3. ZoomInfo (advanced search, help) — find people or companies
  4. Spock (people search on the web, blogs, social networks)
  5. iSearch (review) — can search by name, phone, email, and username (screenname)
  6. Whoozy (review) —- searches the Web and social networks

White page directories

  1. 411 Locate — many different search tools:
  2. WhitePages — lots of different search tools:
    • people & business search
    • reverse phone & address
    • find area codes, zip codes
    • neighborhood search (find the names of people who live near an address)
    • search for an email address
  3. AnyWho — people search, reverse phone
  4. ZabaSearch (advanced search, review) — search by name or by phone number.

Social site search

  1. Wink — people search, phone number; will not only look for the specified name but similar names (i.e. Ted Kennedy, Theodore Kennedy…)
  2. 123people — people search (review, review)
  3. PeekYou — people search, username search
  4. The Internet Address Book — "Find, manage and discover internet addresses worldwide"
  5. Spokeo — good for searching social networking sites; most of the results require that you pay a fee.

23 Project

by dbaodbao (30 Nov 2009 16:50; last edited on 30 Nov 2009 17:36)

Today is mainly a project day, but I noted a few things that aren't written on the site page for today.

Blogs:
Last blog due next Monday so you have time to work on projects.

Projects:
5 more class days to work on projects!
Today: Project Day
Wednesday: Test
Next Monday: Google, Last blogs are due
Next Wednesday: Other miscellaneous search engines
Next Next Monday: Last day of class! Everything for projects is due. December 14!

Test: (WEDNESDAY!)

  • More multiple choice questions: 30/35 instead of 20 (this means more questions total, but most people finished in ~25 minutes last year, so time shouldn't be an issue)
  • Know general framework for class
  • Know what are uses for email alerts
  • Some student questions are too specific - won't need to know all of them
  • Slideshows - posted to individual days

Search Engine Analysis:
If you use a theme where the links don't show up, CHANGE IT!!!!! (Dark Blue is one of these)
Name graphics with a single word.

Rest of today is a project day!

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License