nelson, michael: summarizing archival collections using storytelling techniques
Post on 19-Jan-2017
56 Views
Preview:
TRANSCRIPT
Summarizing archival collections using storytelling techniques
Yasmin AlNoamanyMichele C. WeigleMichael L. Nelson
Old Dominion UniversityWeb Science & Digital Libraries Research Group
www.cs.odu.edu/~mln/@phonedude_mln
Research Funded by IMLS LG-71-15-0077-15
Dodging the Memory Hole Los Angeles, CA, 2016-10-14
2
Archive-It, a subscription-based service, allows creation of collections
> 3,000 collections
~340 institutions
> 10B archived pages
3
Collection title
Collection categorization based on the
curator
Seed URI
Metadata about the collection
Text search box
The group that the resource belongs to
List of the seed
URIs
Timespan of the resource
and the number of times it has been captured
4
Collection understanding and collection summarization are not supported currentlyNot easy to answer “what’s in that collection?” or “how is this collection different from others”?
5
There is more than one collection about “Egyptian Revolution”
• “2010-2011 Arab Spring” https://archive-it.org/collections/3101• “North Africa & the Middle East 2011-2013” https://archive-it.org/collections/2349• “Egypt Revolution and Politics” https://archive-it.org/collections/2358
6
7
8
9
Our early attempts at collection understanding tried to include everything…
“Visualizing digital collections at Archive-It”, JCDL 2012.http://ws-dl.blogspot.com/2012/08/2012-08-10-ms-thesis-visualizing.html
10
1000s of Seeds X 1000s of archived pages == Conventional Vis Methods Not Applicable
11
Idea: Storytelling
12
Stories in literature
Story elements: setting, characters, sequence, exposition, conflict, climax, resolution
Once upon a time
http://www.learner.org/interactives/story/
13
Stories in social media“It's hard to define a story, but I know it when I see it” (Alexander, 2008)
basically, just arranging web pages in time
14
“Storytelling” is becoming a popular technique in social media
15
What are the limitations of storytelling services?
16
The Egyptian Revolution on Storify
17
Bookmarking, not preserving!
18
Despite these limitations, how do we combine storytelling & archives?
19
Use interface people already know how to use to summarize collections
Archived collectionsStorytelling services
Archived enriched stories
20
We sample k mementos from N (k << N) pages of the collection to create a summary story
S1
S2
S3
S4
S2
S1
S3
Collection Y
S3
S2
S1
Collection Z
Archive-It Collections
Collection X
Story
The Web
21
Yasmin hand-crafted stories to summarize the Egyptian Revolution collection for her son, Yousof
https://storify.com/yasmina_anwar/the-egyptian-revolution-on-archive-it-collection
https://storify.com/yasmina_anwar/the-story-of-the-egyptian-revolution-from-archive-
22
How do we generate this automatically?
23
Collections have two dimensions:{Fixed, Sliding} X {Page, Time}
R11
R12
R13
R1n
t1 t3t2 t5t4 tk
…
R21
R22
R23
R2n
…
R31
R32
R33
R3n
…
R41
R42
R43
R4n
…
R51
R52
R53
R5n
…
R61
R62
R63
R6n
…
…
…
…
…
URI
Time
Rk1
Rk2
Rk3
Rkn
…
t6
24
Fixed Page, Fixed Time
A desktop Chrome user-agenthttp://www.cnn.com/2014/02/24/world/africa/egypt-politics/index.html?hpt=wo_c2
Android Chrome user-agenthttp://www.cnn.com/2014/02/24/world/africa/egypt-politics/index.html?hpt=wo_c2
Schneider and McCown, “First Steps in Archiving the Mobile Web: Automated Discovery of Mobile Websites”, JCDL 2013.Kelly et al. “A Method for Identifying Personalized Representations in Web Archives”, D-Lib Magazine 2013 .
25
Feb 1 Feb 1 Feb 2
Feb 4 Feb 5 Feb 7
Feb 9 Feb 11 Feb 11
Fixed Page, Sliding Time
26
Feb. 11, 2011Mubarak resigns Sliding Page, Fixed Time
27
Jan 27 Jan 31
Feb 7Feb 4
Feb 11 Feb 11
Feb 2
Jan 25
Feb 10
Sliding Page, Sliding Time
28
The Dark and Stormy Archives (DSA) framework
Establish a baseline
Reduce the candidate pool of archived pages
Select good representative
pages
Characteristics of human-generated
Stories
Characteristics of Archive-It collections
Exclude duplicates
Exclude off-topic pages
Exclude non-English Language
Dynamically slice the collection
Cluster the pages in each slice
Select high-quality pages from each
cluster
Order pages by time
Visualize
https://pbs.twimg.com/media/BQcpj7ACMAAHRp4.jpg
29
Establish a baseline of social media stories
"Characteristics of Social Media Stories”, TPDL 2015, IJDL 2016.
30
What is the length of a story(the number of resources per story)?
This story has 31 resources
1
3
2
31
What are the types of resources that compose a story?
Quotes
Video
This story has • 19 quotes • 8 images• 4 videos
32
What are the most frequently used domains?
Twitter.com
Twitter.com
Twitter.com
This story has • 90% twitter.com• 7% instagram.com• 3% facebook.com
33
Top 25 domains represents 92% of all domains
34
What differentiates a popular story? (popular = stories with the top 25% of views)
19,795 views 64 views
35
The distributions for the features of the stories
• Based on Kruskal-Wallis test, at the p ≤ 0.05 significance level, the popular and the unpopular stories are different in terms of most of the features
• Popular stories tend to have:• more web elements (medians of 28 vs. 21) • longer timespan (5 hours vs. 2 hours) than the unpopular stories
36
Do popular stories have a lower decay rate?
The 75th percentile of decay rate per popular story is 10% of the resources, while it is 15% in the unpopular stories
37
We found that 28 mementos is a good number for the resources in the stories.
38
Establish a baseline of current Archive-IT collections
"Characteristics of Social Media Stories. What makes a good story?", International Journal on Digital Libraries 2016.
39
The mean and median number of
URIs in a collection
This collection has 435 seed URIs
40
The mean and median number of mementos per URI
This seed URI has 16 mementos
41
The most frequent used domains
abcnews.go.com
blogspot.com
This collection has 30% abcnews.com, 10% blogspot.com, 3% facebook.com
42
Archive-It top 25 is fundamentally different than Storify top 25
43
Archive-It top 25 is fundamentally different than Storify top 25
Twitter is #10 not #1
44
What we archive and what we put in our stories are different
subsets of the web
45
Detecting off-topic pages
"Detecting Off-Topic Pages in Web Archives”, TPDL 2015, IJDL 2016.
46
Archive-It provides their partners with tools that allow them to build themed collections
47
Archive-It tools are about HTTP events / mechanics, not “content”
48
Over 60% of archived versions of hamdeensabahy.com are off-topic
May 13, 2012: The page started as on-topic.
May 24, 2012: Off-topic due to adatabase error.
Mar. 21, 2013: Not working because offinancial problems.
May 21, 2013: On-topic again June 5, 2014: The site has been hacked Oct. 10, 2014: The domain has expired.
http://wayback.archive-it.org/2358/*/http://hamdeensabahy.com
49
How do we automatically detect off-topic pages?
50
We investigated 6 similarity metrics• Textual Content
• cosine similarity of TF-IDF• intersection of the 20 most frequent terms• Jaccard similarity coefficient
• Semantics • Web-based kernel function using a search engine (SE)
• Structural• the change in number of words• the change in content length
51
Textual contentcosine similarity, intersection of the most frequent terms, Jaccard similarity
Method Similaritycosine 0.7TF-Intersection 0.6Jaccard 0.5
52
Textual contentcosine similarity, intersection of the most frequent terms, Jaccard similarity
Method Similaritycosine 0.7TF-Intersection 0.6Jaccard 0.5
Method Similaritycosine 0.0TF-Intersection 0.0Jaccard 0.0
53
Semantics of the textWeb based kernel function using the search engine (SE)
Sahami and Heilman, A Web-based Kernel Function for Measuring the Similarity of Short Text Snippets, WWW 2006
54
Semantics of the textWeb based kernel function using the search engine (SE)
Method SimilaritySE-Kernel 0.7
Sahami and Heilman, A Web-based Kernel Function for Measuring the Similarity of Short Text Snippets, WWW 2006
55
Structural methodsno. of words, content-length
100 109
Method % changeWordCount 0.09
56
Structural methodsno. of words, content-length
100 109
100 5
Method % changeWordCount 0.09
Method % changeWordCount -0.95
57
We built a gold standard data set to evaluate the methods
58
We manually labeled 15,760 mementos
Egypt Revolution and PoliticsURI-Rs: 136URI-Ms: 6,886Off-topic URI-Ms: 384
Occupy MovementURI-Rs: 255URI-Ms: 6,570Off-topic URI-Ms: 458
Columbia Univ. Human Rights collectionURI-Rs: 198URI-Ms: 2,304Off-topic URI-Ms: 94
59
Evaluated 6 methods at 21 thresholds
• Assumed first memento was on-topic
• Combined two methods ('OR') to find best combination method
• 15 combinations• 6,615 tests (15 combinations x 21 thresholds x 21
thresholds)
• Averaged the results at each threshold over the three collections
60
Cosine Similarity performed wellSimilarity Measure Threshold FP FN FP+FN ACC F1 AUC
(Cosine,WordCount) (0.10,-0.85) 24 10 34 0.987 0.906 0.968
(Cosine,SEKernel) (0.10,0.00) 6 35 40 0.990 0.901 0.934
Cosine 0.15 31 22 53 0.983 0.881 0.961
(WordCount,SEKernel) (-0.80,0.00) 14 27 42 0.985 0.818 0.885
WordCount -0.85 6 44 50 0.982 0.806 0.870
SEKernel 0.05 64 83 147 0.965 0.683 0.865
Bytes -0.65 28 133 161 0.962 0.584 0.746
Jaccard 0.05 74 86 159 0.962 0.538 0.809
TF-Intersection 0.00 49 104 153 0.967 0.537 0.740
61
Average precision of 0.89 on 18 Archive-It collections
(Cosine,WordCount) with (0.10,-0.85) thresholds
62
Detecting duplicates in a TimeMap
63
9 mementos for news.egypt.com, but 5 are duplicates
64
How do we dynamically divide the collections into appropriate slices?
65
We expected to see more like this…
The Global Food Crisis collection at Archive-It
66
This is what we found
Egypt Revolution and Politics
Human Rights April 16 Archive Virginia Tech Shooting
Jasmine Revolution 2011 Wikileaks Document Release
67
Selecting representative pages for generating stories(skipping clustering details, but goal is k=28)
68
Quality metrics for selecting mementos• In the DSA, memento quality Mq is calculated as
following: Mq = (1 − wm*Dm) + wql*Sql + wqc*Sqc
• Dm is the memento damage (Brunelle, JCDL 2014)
• Sql is the snippet quality based on the URI level• Sqc is the snippet quality based on URI category• wm, wql, wqc are the weights of memento damage, level,
and category
69
We prefer a higher quality memento (Dm)
http://wayback.archive-it.org/2358/20110201231457/http://news.blogs.cnn.com/category/world/egypt-world-latest-news/
http://wayback.archive-it.org/2358/20110201231622/http://www.bbc.co.uk/news/world/middle_east/
Brunelle et al. Not All Mementos Are Created Equal: Measuring The Impact Of Missing Resources, JCDL 2014
70
We consider the page that gives an attractive snippet
https://wayback.archive-it.org/2358/20110207193404/http://news.blogs.cnn.com/2011/02/07/egypt-crisis-country-to-auction-treasury-bills/
https://wayback.archive-it.org/2358/20110207194425/http://www.cnn.com/2011/WORLD/africa/02/07/egypt.google.executive/index.html?hpt=T1
71
We prefer deep links over high level domains (Sql)
Feb. 11, 2011: the homepage of BBC on Storify
Feb. 11, 2011: the homepage of BBC Middle East section on Storify
Feb. 11, 2011: the article of BBC on Storify
https://wayback.archive-it.org/2358/20110211191429/http://www.bbc.co.uk/
https://wayback.archive-it.org/2358/20110211192204/http://www.bbc.co.uk/news/world-middle-east-12433045
https://wayback.archive-it.org/2358/20110211191942/http://www.bbc.co.uk/news/world/middle_east/
72
Social media pages may not produce good snippets (Sqc)
http://wayback.archive-it.org/1784/20100131023240/http:/twitter.com/Haitifeed/http://wayback.archive-it.org/2358/20141225080305/https:/www.facebook.com/elshaheeed.co.uk
73
Visualizing stories in Storify
74
Remember Yasmin’s hand-crafted stories?
75
Remember Yasmin’s hand-crafted stories?
76
We extract the metadata of the pages and order them chronologically
{ "elements":[ { "permalink":"http://wayback.archive-it.org/694/20070523182134/http://www.usatoday.com/news/nation/2007-04-16-virginia-tech_N.htm", "type":"link", "source":{"href":"http://www.usatoday.com", "name":"www.usatoday.com @ 23, May 2007"} }, { "permalink":"http://wayback.archive-it.org/694/20070530182159/http://www.time.com/time/specials/2007/vatech_victims", "type":"link", "source":{"href":"http://www.time.com", "name":"www.time.com @ 30, May 2007" } }, { "permalink":"http://wayback.archive-it.org/694/20070530182206/http://www.collegiatetimes.com/", "type":"link", "source":{"href":"http://www.collegiatetimes.com", "name":"www.collegiatetimes.com @ 30, May 2007" } }, { "permalink":"http://wayback.archive-it.org/694/20070606234248/http://hokies416.wordpress.com/", "type":"link", "source":{ "href":"http://hokies416.wordpress.com", "name":"hokies416.wordpress.com @ 06, Jun 2007" } }, …{ "permalink":"http://wayback.archive-it.org/694/20070620234329/http://www.hokiesports.com/april16/", "type":"link", "source":{"href":"http://www.hokiesports.com", "name":"www.hokiesports.com @ 20, Jun 2007" } }, ],
"description":"This is an automatically generated story from Archive-It collection.", "title":"April 16 Archive ”}
We override the default metadata to generate more attractive snippets
77
Example of an automatically generated story
Notice the good metadata: images, titles with dates, favicons
78
Evaluating the Dark and Stormy Archive framework
79
What a successful evaluation looks like!
• We use Amazon's Mechanical Turk to compare the following stories:
• Human-generated stories• DSA (automatically) generated stories• Randomly generated stories
• Successful evaluation should result in:• Human and DSA stories are indistinguishable• Human and DSA stories are better than Random
80
Our guidelines for expert archivists at Archive-It for generating stories from the collections
81
We received 23 stories for 10 Archive-It collections
SPST is “Sliding Page, Sliding Time”SPFT is “Sliding Page, Fixed Time” FPST is “Fixed Page, Sliding Time”
82https://storify.com/mturk_exp/3649b1s-57218803f5db94d11030f90b
• Generated by domain experts• Sliding Page, Sliding Time• The Boston Marathon
Bombing collection
83
Automatically generated stories from archived collections
1. Obtain the seed list and the TimeMap of URIs from the front-end interface of Archive- It
2. Extract the HTML of the mementos from the WARC files (locally hosted at ODU) and download the collections that we do not have in the ODU mirror from Archive-It
3. Extract the text of the page using the Boilerpipe library 4. Eliminate the off-topic pages based on the best-performing method ((Cosine,
Word-Count) with the suggested thresholds (0.1, −0.85))5. Exclude the duplicates of each TimeMap 6. Eliminate the non-English language pages7. Slice the collection dynamically and then cluster the mementos of each slice
using DBSCAN algorithm8. Apply the quality metrics to select the best representative pages9. Sort the selected mementos chronologically then put them and their metadata
in a JSON object
84https://storify.com/mturk_exp/3649b0s
• Automatically generated story • Sliding Page, Sliding Time• The Boston Marathon
Bombing collection
85
Random stories
28 mementos were randomly selected from each collection before excluding off-topic and duplicate pages
86https://storify.com/mturk_exp/3649b2s-57227227bb79 048c2d0388dc
• Randomly generated story• Sliding Page, Sliding Time• The Boston Marathon
Bombing collection
87https://storify.com/mturk_exp/3649bads
if someone prefers this story, we exclude their results
• Poorly generated story• Sliding Page, Sliding Time• The Boston Marathon
Bombing collection
88
MT experiment setup
• Three HITs for each story (69 HITs to evaluate 23 stories); two comparisons per HIT:
• HIT1: human vs. automatic, human vs. poor• HIT2: human vs. random, human vs. poor• HIT3: random vs. automatic, automatic vs. poor
• 15 distinct turkers with master (have high acceptance rate) qualification for each HIT
• We rejected the submissions contained poorly-generated stories and the HITs that were completed in less than 10 seconds (mean time per HIT = 7 minutes)
• 989 out of 1,035 (69*15) valid HITs
• We awarded the turker $0.50 per HIT
https://www.mturk.com/mturk/help?helpPage=worker#what_is_master_worker
89
A sample HIT
90
DSA == Human(Human,DSA) > Random
91
Automatic versus Human
Sliding Page, Sliding Time Sliding Page, Fixed Time Fixed Page, Sliding Time
92
Human versus Random
Sliding Page, Sliding Time Sliding Page, Fixed Time Fixed Page, Sliding Time
93
Automatic versus Random
Sliding Page, Sliding Time Sliding Page, Fixed Time Fixed Page, Sliding Time
94
Success!
DSA-generated stories are just as good as stories generated by human experts
95
Use interface people already know how to use to summarize collections
Archived collectionsStorytelling services
Archived enriched stories
All the code, datasets, papers, slides, etc.:http://bit.ly/YasminPhD
top related