on enhancing the user experience in web search engines
DESCRIPTION
On Enhancing the User Experience in Web Search Engines. Franco Maria Nardini. About Me. I joined the HPC Lab in 2006 Master Thesis Ph.D. in 2011, University of Pisa Thesis: “Query Log Mining to Enhance User Experience in Search Engines” m ail: [email protected] - PowerPoint PPT PresentationTRANSCRIPT
On Enhancing the User Experiencein Web Search Engines
Franco Maria Nardini
About Me
• I joined the HPC Lab in 2006– Master Thesis
• Ph.D. in 2011, University of Pisa– Thesis: “Query Log Mining to Enhance User
Experience in Search Engines”
• mail: [email protected]• web: http://hpc.isti.cnr.it/~nardini• skype: francomaria.nardini
Query Suggestion
with Daniele Broccolo, Lorenzo MarconRaffaele Perego, Fabrizio Silvestri
Our Contribution: Search Shortcuts
Our Contribution: Search Shortcuts
Our Contribution: Search Shortcuts
• Search Shortcuts:– It uses the “happy ending” stories in the query log
to help new users;• Efficient:– All the “stuff” is stored on a inverted index:
retrieval problem;• Effective: (head, torso, tail)– New evaluation methodology confirming this
evidencies: TREC Diversity Track.Daniele Broccolo, Lorenzo Marcon, Franco Maria Nardini, Fabrizio Silvestri, Raffaele Perego, Generating Suggestions for Queries in the Long Tail with an Inverted Index, IP&M, 2011.
Some Results
What’s Next?!• Why not to use Machine Learning?– Machine learning is helping a lot in the IR
community;– Better and “fine-graned” ranking as it could take
into account important signals that are not fully-exploited nowadays;
– It may helps in filtering redundant suggestions and choosing the “best” expressive ones (for each intent).
under exploration withMarcin Sydow (PJIIT),
Raffaele Perego, Fabrizio Silvestri
Signals
• Which signals we would like to capture?– Relevance to the given query;– Diversity with respect to a subtopic list;– Serendipity of suggestions;– Novelty with respect to news/trends on Twitter;
• How do we catch them?• How do we combine them?• The “training” set is a problem.
Query Suggestion: Ranking
• A two-step architecture– First step to produce a list of candidates;– Second step as a ML architecture composed of two
different (cascade) stages of ranking:• First round to rank suggestions w.r.t. the query;• Second round to understand “diversity”.
Diversification ofWeb Search Engine Results
withGabriele Capannini, Raffaele Perego, Fabrizio Silvestri
Our Contribution
• We design a method for efficiently diversify results from Web search engines.– Same effectiveness of other state-of-the-art
approaches;– Extremely fast in doing the “hard” work;
• Intents behind “ambiguous” queries are mined from query logs;
Capannini G., Nardini F.M., Silvestri F., Perego R., A Search Architecture Enabling Efficient Diversification of Search Results, Proc. DDR Workshop 2011.Capannini G., Nardini F.M., Silvestri F., Perego R., Efficient Diversification of Web Search Results. Proceedings of VLDB 2011 (PVLDB), Volume 4, Issue 7.
Our Contribution
Our Contribution
Some Results
What’s Next?• A modern ranking architecture:– Effective:• Users should be happy of the results they receive;
– Efficient:• Low response times (< 0.1 s);
– Easy to adapt:• Continuous crawling from the Web;• Continuous users’ feedback;
with Berkant Barla Cambazoglu (Yahoo! Barcelona),
Gabriele Capannini, Raffaele Perego, Fabrizio Silvestri
Let’s Plug All Together
BM25 Scorer1 … Scorern
Query
Index
Second Phase
First Phase
Results
Scorerdiv
SS
• A way for efficiently diversifying “ambiguous” queries;• SS teaches how to “diversify” the current user query;• Scorerdiv computes the diversity “signal” of each document and
rerank the final results list;
Possible intents behind the query
Retrieval over Query Sessions
with M-Dyaa AlBakour
(University of Glasgow)
Main Goals
• Question 1)– Can Web search engines improve their
performance by using previous user interactions? (including previous queries, clicks on ranked results, dwell times, etc.)
• Question 2)– How do we evaluate system performance over an
entire query session instead of a single query?
TREC Session Track• Two editions of the challenge: 2010, 2011– query, previous queries;– urls + docs, urls + docs + dwell time;– Two different evaluations: last subtop., all subtop.
• “Query expansion” with Search Shortcuts:– weighted by means of user interaction data;– “history-based” recommendation;
• Follow-up with tuning of the parameters.Ibrahim Adeyanju, Franco Maria Nardini, M-Dyaa Albakour, Dawei Song, Udo Kruschwitz, RGU-ISTI-Essex at TREC 2011 Session Track, TREC Conference, 2011.Franco Maria Nardini, M-Dyaa Albakour, Ibrahim Adeyanju, Udo Kruschwitz, Studying Search Shortcuts in a Query Log to Improve Retrieval Over Query Sessions, SIR 2012 in conjunction with ECIR 2012.
Some Results
What’s Next?• Entity-based representation of the user
session.– to reduce the “sparsity” of the space.
Challenges
• How those systems really affect (and modify) the behavior of the user?– Is it possible to quantify it? (metrics?)– What do we need to observe?
• Toward the “perfect result page”:– accurate models for blending different sources of
results.
Little Announcement
http://tf.isti.cnr.it
• Models and Techniques for Tourist Facilities• Evaluation and Test Collections• User Interaction and Interfaces
Paper Deadline
06/25/2012
Questions!?!