Jaume Nualart | Universitat Oberta de Catalunya (original) (raw)

Papers by Jaume Nualart

Good readability of text is important to ensure efficiency in communication and eliminate risks o... more Good readability of text is important to ensure efficiency in communication and eliminate risks of misunderstanding. Patent claims are an example of text whose readability is often poor. In this paper, we aim to improve claim readability by a clearer presentation of its content. Our approach consist in segmenting the original claim content at two levels. First, an entire claim is segmented to the components of preamble, transitional phrase and body, using a rule-based approach. Second, a conditional random field is trained to segment the components into clauses. An alternative approach would have been to modify the claim content which is, however, prone to also changing the meaning of this legal text. For both segmentation levels, we report results from statistical evaluation of segmentation performance. In addition, a qualitative error analysis was performed to understand the problems underlying the clause segmentation task. Our accuracy in detecting the beginning and end of preamble text is 1.00 and 0.97, respectively. For the transitional phase, these numbers are 0.94 and 1.00 and for the body text, 1.00 and 1.00. Our precision and recall in the clause segmentation are 0.77 and 0.76, respectively. The results give evidence for the feasibility of automated claim and clause segmentation, which may help not only inventors, researchers, and other laypeople to understand patents but also patent experts to avoid future legal cost due to litigations.

Information Research, 2013

ABSTRACT

International Journal of Information Management, Oct 1, 2015

Journal and digital library portals are the information systems that researchers turn to most fre... more Journal and digital library portals are the information systems that researchers turn to most frequently for undertaking and disseminating their academic work. However their interfaces have not been improved. We propose an articulation of the navigation and search systems in a single visual solution that would allow the simultaneous exploration and interrogation of the information system. Area is a low-cost visualization tool that is easy to implement, and which can be used with large collections of documents. Moreover, it has a short learning curve that enhances both user-experience and user-satisfaction with journal and digital library websites.

Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They a... more Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They are also important tools in patient empowerment, but a patient's comprehension of the information is often suboptimal. Continuing in the tradition of focusing on automated approaches to increasing patient comprehension, The CLEFeHealth2014 lab tasked participants to visualize the information in discharge summaries while also providing connections to additional online information. Participants were provided with six cases containing a discharge summary, patient profile and information needs. Of fifty registrations, only the FLPolytech team completed all requirements related to the task. They augmented the discharge summary by linking to external resources, inserting structure related to timing of the information need (past, present future), enriching the content, i.e., with definitions, and providing meta-information, e.g., how to make future appointments. Four panellists evaluated the submission. Overall, they were positive about the enhancements, but all agreed that additional visualization could further improve the provided solution.

Good readability of text is important to ensure efficiency in communication and eliminate risks o... more Good readability of text is important to ensure efficiency in communication and eliminate risks of misunderstanding. Patent claims are an example of text whose readability is often poor. In this paper, we aim to improve claim readability by a clearer presentation. This segments the original claim text first to components of the preamble, transition, and body text and then the components further to clauses. An alternative approach would have been to modify the claim content which is, how- ever, prone to also changing the mean- ing of this legal text. Our rule-based method detects the beginning and end of the preamble (transition) [body text] with the accuracy of 100% and 97% (94% & 100%) [100% & 100%], respectively. In clause segmentation, our conditional ran- dom field (punctuation and keyword-based baseline) has the precision of 77% (41%) and recall of 76% (29%). The most com- mon reasons for segmentation errors are ambiguous coordinating conjunctions and consecutive segmentation keywords. The results give evidence for the feasibility of automated claim and clause segmentation, which may help not only inventors, re- searchers, and other laypeople to under- stand patents but also patent experts to avoid future legal cost due to litigations.

HAL (Le Centre pour la Communication Scientifique Directe), Sep 15, 2014

This work introduces Diggersdiaries, an web interface to a historical textual document collection... more This work introduces Diggersdiaries, an web interface to a historical textual document collection. Digital collections are rich in content, but traditional search-based and faceted-based interfaces cannot represent their richness efficiently - e.g. for time-poor and casual browsing. This project addresses this challenge using data analysis (topic models) transparently integrated into reading-centric interface as a two-level browsing menu of semantic topics. The interface offers multiple exploration visualization tools. Its main contribution it is that the interface is fully reading-oriented. The tool is available at http://diggersdiaries.org

PhD Thesis - University of Canberra (ACT - Australia), 2016

This research brings together data analysis with software engineering and visual- isation, with ... more This research brings together data analysis with software engineering and visual-
isation, with a specific focus on text mining and large document collections. My
aim is to devise new, rich, and simple visualisation interfaces, which I call deep
interfaces.
With deep interfaces I introduce the idea-rich content as a product of the stat-
istical analysis combined with human curation of labels and interpreted as a flow
of subjectivity, complexity, and diversity between reader and interface and vice
versa.
The focus of such interfaces is not the representation of textual document col-
lections as in Moretti’s distant reading, but to revisit traditional reading from the
point of view of state of the art methods of textual analysis. Thus, the proposed
interfaces can help us discover and explore text document collections by reading
their contents. This is a practice-led research project that develops theoretical
issues through the generation of practical artefacts. The research process is cu-
mulative, following a reflexive methodology. The key outcomes of the project are
embodied in an interface to a large collection of ANZAC war diaries: Diggers’
Diaries — http://diggersdiaries.org.

Proceedings of the 3rd Workshop on Predicting and Improving Text Readability for Target Reader Populations (PITR), 2014

Abstract. Discharge summaries serve a variety of aims, ranging from clinical care to legal purpos... more Abstract. Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They are also important tools in patient empowerment, but a patient’s comprehension of the information is often suboptimal. Continu-ing in the tradition of focusing on automated approaches to increasing patient comprehension, The CLEFeHealth2014 lab tasked participants to visualize the information in discharge summaries while also providing connections to addi-tional online information. Participants were provided with six cases containing a discharge summary, patient profile and information needs. Of fifty registra-tions, only the FLPolytech team completed all requirements related to the task. They augmented the discharge summary by linking to external resources, insert-ing structure related to timing of the information need (past, present future), en-riching the content, i.e., with definitions, and providing meta-information, e.g., how to make future appointments. Four panellists ev...

This research project aims to improve the way humans work with textual documents when doing tasks... more This research project aims to improve the way humans work with textual documents when doing tasks such as exploring, discovering, searching, filtering, collecting, indexing, comparing, or just reading. This research project studies theoretical foundations and practical uses of text visualization techniques. As a contribution to theory, the research project presents a classification schema for text visualization approaches based on visual features instead of task-solving capabilities (Paper I). As a contribution to practice, how to improve interfaces of digital libraries (DL) has been studied. Two practical proposals for approaching text are introduced: one for single text representation, called Texty (Paper II), and another for text collections exploration and overview, called Area (Paper III). This research project discusses the contrast between the growing popularity of text visualization, presented as a subfield of data visualization, and the lack and urgency, nowadays, of intera...

Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They a... more Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They are also important tools in patient empowerment, but a patient's comprehension of the information is often suboptimal. Continu- ing in the tradition of focusing on automated approaches to increasing patient comprehension, The CLEFeHealth2014 lab tasked participants to visualize the information in discharge summaries while also providing connections to addi- tional online information. Participants were provided with six cases containing a discharge summary, patient profile and information needs. Of fifty registra- tions, only the FLPolytech team completed all requirements related to the task. They augmented the discharge summary by linking to external resources, insert- ing structure related to timing of the information need (past, present future), en- riching the content, i.e., with definitions, and providing meta-information, e.g., how to make future appointments. Four panellists eva...

We present Crossreads, a manner to deconstruct linear narrative text in order to read text in mul... more We present Crossreads, a manner to deconstruct linear narrative text in order to read text in multiple orders. This is an ongoing project aims to study data multiplicity, as well as textual visualization interfaces. The process starts with the selection of a text, which is later segmented into small blocks, and the textual similarity among them is calculated, forming a network data set. Finally, a web interface allows the user to explore and read through the created network of text.

EACL 2014, 2014

Good readability of text is important to ensure efficiency in communication and eliminate risks... more Good readability of text is important
to ensure efficiency in communication
and eliminate risks of misunderstanding.
Patent claims are an example of text whose
readability is often poor. In this paper,
we aim to improve claim readability by
a clearer presentation. This segments the
original claim text first to components of
the preamble, transition, and body text and
then the components further to clauses. An
alternative approach would have been to
modify the claim content which is, how-
ever, prone to also changing the mean-
ing of this legal text. Our rule-based
method detects the beginning and end of
the preamble (transition) [body text] with
the accuracy of 100% and 97% (94% &
100%) [100% & 100%], respectively. In
clause segmentation, our conditional ran-
dom field (punctuation and keyword-based
baseline) has the precision of 77% (41%)
and recall of 76% (29%). The most com-
mon reasons for segmentation errors are
ambiguous coordinating conjunctions and
consecutive segmentation keywords. The
results give evidence for the feasibility of
automated claim and clause segmentation,
which may help not only inventors, re-
searchers, and other laypeople to under-
stand patents but also patent experts to
avoid future legal cost due to litigations.

Profesional De La Informacion, 2014

His work has focused on information architecture and visualization. He is author of the book Arqu... more His work has focused on information architecture and visualization. He is author of the book Arquitectura de la información en entornos web (Trea, 2010).

Profesional De La Informacion, 2014

Information Research, 2013

ABSTRACT

International Journal of Information Management, Oct 1, 2015

Good readability of text is important to ensure efficiency in communication and eliminate risks o... more Good readability of text is important to ensure efficiency in communication and eliminate risks of misunderstanding. Patent claims are an example of text whose readability is often poor. In this paper, we aim to improve claim readability by a clearer presentation. This segments the original claim text first to components of the preamble, transition, and body text and then the components further to clauses. An alternative approach would have been to modify the claim content which is, how- ever, prone to also changing the mean- ing of this legal text. Our rule-based method detects the beginning and end of the preamble (transition) [body text] with the accuracy of 100% and 97% (94% & 100%) [100% & 100%], respectively. In clause segmentation, our conditional ran- dom field (punctuation and keyword-based baseline) has the precision of 77% (41%) and recall of 76% (29%). The most com- mon reasons for segmentation errors are ambiguous coordinating conjunctions and consecutive segmentation keywords. The results give evidence for the feasibility of automated claim and clause segmentation, which may help not only inventors, re- searchers, and other laypeople to under- stand patents but also patent experts to avoid future legal cost due to litigations.

HAL (Le Centre pour la Communication Scientifique Directe), Sep 15, 2014

PhD Thesis - University of Canberra (ACT - Australia), 2016

Proceedings of the 3rd Workshop on Predicting and Improving Text Readability for Target Reader Populations (PITR), 2014

Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They a... more Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They are also important tools in patient empowerment, but a patient's comprehension of the information is often suboptimal. Continu- ing in the tradition of focusing on automated approaches to increasing patient comprehension, The CLEFeHealth2014 lab tasked participants to visualize the information in discharge summaries while also providing connections to addi- tional online information. Participants were provided with six cases containing a discharge summary, patient profile and information needs. Of fifty registra- tions, only the FLPolytech team completed all requirements related to the task. They augmented the discharge summary by linking to external resources, insert- ing structure related to timing of the information need (past, present future), en- riching the content, i.e., with definitions, and providing meta-information, e.g., how to make future appointments. Four panellists eva...

EACL 2014, 2014

Profesional De La Informacion, 2014

Figure 1: Detail of Diggersdiaries interface: the two-level menu of topics support reading and ex... more Figure 1: Detail of Diggersdiaries interface: the two-level menu of topics support reading and exploration of a large collection of World War I ANZAC soldiers' diaries. Abstract This work introduces Diggersdiaries, an web interface to a historical textual document collection. Digital collections are rich in content, but traditional search-based and faceted-based interfaces cannot represent their richness efficiently-e.g. for time-poor and casual browsing. This project addresses this challenge using data analysis (topic models) transparently integrated into reading-centric interface as a two-level browsing menu of semantic topics. The interface offers multiple exploration visualization tools. Its main contribution it is that the interface is fully reading-oriented. The tool is available at http://diggersdiaries.org

We present Crossreads, a manner to deconstruct linear nar- rative text in order to read text in ... more We present Crossreads, a manner to deconstruct linear nar-
rative text in order to read text in multiple orders. This is
an ongoing project aims to study data multiplicity, as well
as textual visualization interfaces. The process starts with
the selection of a text, which is later segmented into small
blocks, and the textual similarity among them is calculated,
forming a network data set. Finally, a web interface allows
the user to explore and read through the created network of
text.

Improving the claim readability through content presentation rather than modification Motivati... more Improving the claim readability through content presentation rather than modification

Motivation
- Extremely long sentences. Description of complex technical issues. Legal terminology
- Democratization of invention described in patents, and thereof, of human knowledge.

Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They a... more Discharge summaries serve a variety of aims, ranging from clinical care to legal purposes. They are also important tools in patient empowerment, but a patient’s comprehension of the information is often suboptimal. Continuing in the tradition of focusing on automated approaches to increasing patient comprehension, The CLEFeHealth2014 lab tasked participants to visualize the information in discharge summaries while also providing connections to additional online information. Participants were provided with six cases containing a discharge summary, patient profile and information needs. Of fifty registrations, only the FLPolytech team completed all requirements related to the task. They augmented the discharge summary by linking to external resources, inserting structure related to timing of the information need (past, present future), enriching the content, i.e., with definitions, and providing meta-information, e.g., how to make future appointments. Four panellists evaluated the submission. Overall, they were positive about the enhancements, but all agreed that additional visualization could further improve the provided solution.

Motivation Most of conferences proceedings present their content as a one-dimension, non-intera... more Motivation
Most of conferences proceedings present their content as a
one-dimension, non-interactive list of papers on a web page.
However, the reader of this kind of presentation might not
know the reason for the paper order; does not get an
overview of the contents or relations between the papers; and
has very limited search and filtering functionalities available.

Aim
To explore more effective interfaces to represent contents of
conference proceedings. One of the inspiring works in this
direction is called Word Storms, by Castella and Sutton
(2013), applied to the International Conference on Machine
Learning , ICML 2012 (1)(2)

The data
In collaboration with Mark Reid, we used the list of accepted papers from JMLR Workshop and Conference Proceedings Volume 28 : Proceedings of The 30th International Conference on Machine Learning.This is a collection of 282 papers.

The analysis
Wray Buntine conducted the analysis using topic models. Firstly we created a collection of representative texts of ML (from books to Arxiv papers). From this analysis, we created ten topic and, instead of topic1, topic2, topic3, we gave a human name to each of them.

Finally every paper from JMLR dataset has being scored according to the ten topics.

http://research.nualart.cat/area-feminicides/ Violence against women is one of the main threats ... more http://research.nualart.cat/area-feminicides/
Violence against women is one of the main threats that women have to face in our societies.Feminicides are the most extreme and irreversible form of this violence. It is the violence resulting in death where gender is a determining variable in explaining the crime.
The visualizations of Feminides in AREA were created in 2006. This project aimed at collecting existing information on feminicides in the Spanish State to create interactive visualizations to reach awareness about the extent of violence against women in our society. Feminicides in AREA show violence against women that resulted in death from 2000 to 2010 in the Spanish State in a simple and disturbing way.