=Paper=
{{Paper
|id=Vol-2068/preface-esida
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-2068/preface-esida.pdf
|volume=Vol-2068
}}
==None==
https://ceur-ws.org/Vol-2068/preface-esida.pdf
Second Workshop on Exploratory Search and Interactive
Data Analytics (ESIDA)
Dorota Głowacka? Evangelos Milios◦ Axel J. Soto4 Fernando V. Paulovich◦ Denis ParraO
? ◦ 4 O
University of Edinburgh Dalhousie University ICIC-CONICET PUC Chile
? ◦ ◦
dorota.glowacka@ed.ac.uk, eem@cs.dal.ca, paulovich@dal.ca,
4
axel.soto@cs.uns.edu.ar, O dparra@ing.puc.cl
ABSTRACT Despite an increasing interest in exploratory search and in-
This is the second edition of the Workshop on Exploratory teractive data analytics [6], several research questions still re-
Search and Interactive Data Analytics (ESIDA). This series main open, e.g.:
of workshops emerged as a response to the growing inter-
• What is the scope of exploratory search [10]?
est in developing new methods and systems that allow users
• How to design systems that can support both traditional
to interactively explore large volumes of data, such as docu-
lookup search and exploratory search [2] as well as support
ments, multimedia or specialised collections, such as biomed-
personalisation [9]?
ical datasets. There are various approaches to supporting
• How machine learning methods can be complemented with
users in this interactive environment ranging from the devel-
user domain knowledge to improve information finding or
opment of new algorithms through visualisation methods to
sense making [4]?
analysing users’ search patterns. The overarching goal of ES-
• How can exploratory search and interactive analytics be
IDA is to bring together researchers working in areas that
used in specialised domains, such as bioinformatics [13],
span across multiple facets of exploratory search and data
mobile network data [7], or social media [5]?
analytics to discuss and outline research challenges for this
novel area.
WORKSHOP OVERVIEW
The aim of the workshop was to investigate different aspects
Author Keywords of the design and evaluation of interactive search and data
exploratory search, interactive search, data analytics, analytics algorithms and systems that tightly couple interac-
personalisation tive visualization with analytics of the content of the data sets.
The workshop included two keynotes, four regular papers and
ACM Classification Keywords
one short paper.
H.5.m. Information Interfaces and Presentation (e.g. HCI): The first keynote, by Giulio Jacucci (University of Helsinki,
Miscellaneous Finland), was titled “Entity Based User Interfaces for Re-
sourceful Information Exploration”. The talk proposed a
new approach characterised by entity modelling of informa-
BACKGROUND tion, conversational and synergic human-computer interac-
While retrieval techniques operating on text or semantic tion, combining search and recommendation, and mapping
annotations have become an industry standard, traditional relationships and representations of intent, queries and re-
search by keyword query becomes cumbersome for other sults. To demonstrate these features the talk discussed re-
forms of data (e.g. images, video, music) and even for textual cent systems developed by the speaker’s team including an
data in the case of ambiguous queries (e.g. bank referring to interactive model of intent represented by keywords used to
side of a river or a financial institution). Also, in complex direct exploratory search, a system utilising entity based par-
scenarios, or when the path from data to decision is not clear, allel search streams, and an interactive map visualisation for
exploratory search is necessary to understand and seek infor- multi-aspect information retrieval. The talk summarised the
mation interactively. Therefore, by actively engaging the user principles embodied in the presented system towards a frame-
in the information retrieval loop, the user can explore a given work that considers principle of how to employ entities in in-
dataset more easily as well as gradually direct their search to formation exploration.
a more specific area of the search space.
The second keynote, by Shlomo Berkovsky (CSIRO, Aus-
tralia), was titled “Improving interactive information discov-
ery”. The talk presented computational ways for improving
the interaction of users with the recommendation lists or Web
search results, aiming to devise a method that simplifies and
shortens information discovery for users. One practical appli-
cation of a generalised linear search (combination of the es-
tablished linear and generalised search approaches) was pre-
©2018. Copyright for the individual papers remains with the authors. Copying per-
mitted for private and academic purposes. sented, evaluated both with offline datasets and in an online
ESIDA’18, March 11, 2018, Tokyo, Japan
user study. It was discussed how similar interactive informa- 5. A. Dang, A. Moh’d, A. Gruzd, E. Milios, and
tion discovery approaches can be applied to scenarios beyond R. Minghim. An offline–online visual framework for
recommender systems and Web search. clustering memes in social media. In From Social Data
Mining and Analysis to Prediction and Community
Sense making of document collections was the subject of Detection, pages 1–29. Springer, 2017.
two papers. Hierarchically organized social media discus-
sions were visualized as scatter plots via dimensionality re- 6. D. Glowacka, E. Milios, A. J. Soto, and F. Paulovich.
duction [11]. A variety of interaction techniques was demon- Exploratory search and interactive data analytics. In
strated to help reveal semantic relationships between posts Proceedings of the 22Nd International Conference on
that are not associated with the given hierarchy. Interactive Intelligent User Interfaces Companion, IUI ’17
topic models were used for highlighting key points of dis- Companion, pages 9–11, New York, NY, USA, 2017.
cussion in online petition data [3]. Content curation of the ACM.
contents of a social Q&A system, by performing interactive
7. E. Graells-Garrido, D. Caro, and D. Parra. Toward
cluster analysis to cluster together near-duplicate entries to
finding latent cities with non-negative matrix
improve search relevance was the focus of one paper [12].
factorization. arXiv preprint arXiv:1801.09093, 2018.
Detection and classification of sentiment, aiming to identify
outliers and fake news in news corpora is addressed in [8], 8. C. G. Harris. Searching for diverse perspectives in news
using a Recursive Neural Network, in particular Long Short articles: Using an lstm network to classify sentiment. In
Term Memory (LSTM) network. An interactive query inter- ESIDA, 2018.
face supported an empirical study in which users classify sen-
9. A. Medlar, J. Pyykkö, and D. Glowacka. Towards
timent on name entities. User input is compared with the re-
fine-grained adaptation of exploration/exploitation in
sult obtained from the LSTM network. The last two papers
information retrieval. In Proceedings of the 22Nd
focus on human annotation of documents to train machine
International Conference on Intelligent User Interfaces,
learning systems. A user interface facilitates and optimizes
IUI ’17, pages 623–627, New York, NY, USA, 2017.
the interactive steps of document presentation, inspection and
ACM.
labelling of publications, and supports the incorporation of
domain-specific features (terminology gazetteers and docu- 10. E. Palagi, F. Gandon, A. Giboin, and R. Troncy. A
ment metadata), in [1]. survey of definitions and models of exploratory search.
In Proceedings of the 2017 ACM Workshop on
REFERENCES Exploratory Search and Interactive Data Analytics,
1. C. Aguirre, S. Coen, M. F. D. L. Torre, W. Hsu, and ESIDA ’17, pages 3–8, New York, NY, USA, 2017.
M. Rys. Towards faster annotation interfaces for ACM.
learning to filter in information extraction and search. In
ESIDA, 2018. 11. J. Peltonen, Z. Lin, K. Järvelin, and J. Nummenmaa.
PIHVI: Online forum posting analysis with interactive
2. K. Athukorala, A. Medlar, A. Oulasvirta, G. Jacucci, hierarchical visualization. In ESIDA, 2018.
and D. Glowacka. Beyond relevance: Adapting
exploration/exploitation in information retrieval. In 12. I. Podgorny and C. Gielow. Semi-automated prevention
Proceedings of the 21st International Conference on and curation of duplicate content in social support
Intelligent User Interfaces, IUI ’16, pages 359–369, systems. In ESIDA, 2018.
New York, NY, USA, 2016. ACM.
13. A. J. Soto, C. Zerva, R. Batista-Navarro, and
3. G. Cai and F. Sun. Interactive visualization for topic S. Ananiadou. Litpathexplorer: a confidence-based
model curation. In ESIDA, 2018. visual text analytics tool for exploring
4. D. B. Coimbra, R. M. Martins, T. T. Neves, A. C. Telea, literature-enriched pathway models. Bioinformatics, 1:9,
and F. V. Paulovich. Explaining three-dimensional 2017.
dimensionality reduction plots. Information
Visualization, 15(2):154–172, 2016.