Paloma Cáceres

Model transformations for hypertext modeling on web information systems

In response to the lack of suitable methods to build the navigation model of Web Information Systems (WIS), we presented in past works HM3, the Hypertext Modeling Method of MIDAS, a methodological framework for the agile development of WIS. We proposed a method to obtain the navigation model of a WIS, starting from the conceptual data model and the user requirements, collected in the use case model. In this work, continuing with the MDA approach for WIS development that we are following with HM3, we define the model to model transformations needed at the PIM level to get the hypertext model from the behavioral and content models. Firstly we specify the transformation rules with natural language to later map them to graph transformation rules.

A User Service Oriented Method to Model Web Information Systems

Nowadays it is possible to develop Web Information Systems (WIS) that allow us to buy a fly ticket or to book a hotel, as well as to get specific information through the Internet. However, due to the lack of suitable methods to build the navigation model of WIS, frequently the users are lost while navigating through the system, because they do not know how to use it. Traditional methodologies for WIS development usually propose to obtain the navigation model from the conceptual data model, without taking into account the behavior of systems. Unlike these methodologies, we propose to address the problem of navigation model construction from a user service oriented perspective. In this work we explain, through a case study, the hypertext modeling method proposed by MIDAS. It allows to build more intuitive and user friendly WIS. Moreover, we present a comparative study in which we remark the main benefits and advantages of our method with regard to the traditional methodologies for hypertext modeling.

MIDAS/BD: A Methodological Framework for Web Database Design

One of the most important trends in Information System research is concerning Web application development. Generally, Web Information Systems are developed ad-hoc, that is, without modeling them before their implementation. At best, developers use techniques and methodologies directly imported from the traditional Software Engineering field. However, classic methodologies do not fit to some key aspects of Web applications as, for example, navigation design. This is the reason why different modeling techniques and methodologies for Web Information Systems development have appeared. MIDAS is a methodological framework for Web Information System development that can be customized according to the type of application we want to develop (hypermedia applications, database applications, etc.). In this paper we present MIDAS/BD, the specific part of MIDAS for Web Database development. MIDAS/BD is based on XML and (object -) relational technology.

Participation of women in software engineering publications

To the best of our knowledge, no works analyzing the participation of women as authors and editors in software engineering research publications currently exist. We have therefore followed a well-defined procedure in order to conduct an empirical study of female participation in 12 leading software engineering journals. We have analyzed the gender of the authors, editorial board members, associate editors and editors-in-chief over a two-year period in order to analyze, on the one hand, the rate of participation of women as authors and as editors in software engineering publications, and on the other, whether women are underrepresented. We have also analyzed the female distribution of authors and editors according to the geographical location of their institutions. This was done by first selecting the journals to be used as the population for data collection which then allowed us to identify female authors of papers and female editors, including the country in which their institutions are located. This eventually led to an analysis of female participation in order to understand representation rates. We analyzed 3,546 authors of 1,266 papers in 61 different countries, and 363 members of editorial boards in 30 different countries. The results of this analysis provide quantitative data concerning the participation of women as authors and editors in major software engineering journals including their distribution per country, in which important differences have been found. The results obtained were first used to compare the participation of women as authors and editors and were then used to carry out a series of simulations in order to be able to statistically confirm whether women are underrepresented. The study shows, amongst other things, that women are not underrepresented as editorial boards members and as editors-in-chief of the journals studied, although their representation as editors-in-chief is low.

Computer science research: more production, less productivity

It is commonly accepted that scientific research or, more precisely, the number of scientific publications, in computer science has greatly increased over the last few years. The reason would appear to be the pressure to publish, coined by the expression ”Publish or perish”, which is, among other things, necessary for promotions and applications for grants or projects. In this paper we have conducted a study that covers computer science publications from 1936 to 2010 in order to quantify this increase in publications regarding computing research. We have considered the computing conferences and journals available in the DBLP computer science bibliography (DBLP 2013) database, including more than 1.5 million papers, and more than 4 million authors (more than 900,000 different people), corresponding to about 1,000 different journals and 3,000 different conferences and workshops. Our study confirms and quantifies these increases with regard to the number of papers, number of authors, number of papers per author, etc. However, it also reaches a surprising conclusion: the real productivity of researchers has decreased throughout history. The reason for this decrease is the average number of authors per paper, which has grown significantly and is currently three.

Navigation modelling from a user services oriented approach

Traditional methodologies for Web Information Systems (WIS) development usually propose to get the navigation model from the conceptual data model. Unlike these methodologies, that follow a structural approach to build the navigation model, we propose to address the problem of navigation model construction from a user services oriented perspective. The method is based on identifying conceptual user services, that is specific services required by the user, and all the models are user services oriented. In this way, the navigation model is focused on the services required by the user. It also identifies just one specific route for each conceptual user service that guides the navigation of the user through the WIS. In this paper we explain the method and we show, by a case study, how the navigation model conceived from a user services oriented approach is more intuitive and user friendly.

Precise wis development

In recent years, Internet has become the platform that supports most areas in organizations. This fact has led to the appearance of specific tools for the construction of Web Information Systems (WIS). However, in these tools an absence of functionalities for verification and validation (V&V) of the models built has been detected. This work aims to redress this absence with the definition of a strategy for the specification of the models used in the WIS development that can be used with V&V objectives. The approach has been validated in a specific methodology (MIDAS) and an associated tool (MIDAS-CASE), both aligned with the MDA proposal. One of the diagrams used in this methodology has been formalized: the extended navigation model, achieving a precise specification of this diagram. This specification permits the definition and formal verification of properties related to this diagram, as well as its validation. The specification and verification of a concrete property are also shown.

Improving usability in e-democracy systems: systematic development of navigation in an e-participatory budget system

E-democracy refers to a set of methods to address social issues through the web, possibly by using decision support methodologies to involve citizens in decision-making processes. Participatory budgets, in which citizens participate in budget allocation, are an example. Web applications supporting these processes entail complex interactions in which users communicate their preferences, negotiate alternatives with other participants, etc. Thus, developing web applications may determine the usability and success of an application. To better reflect the paths which may drive users through an application, navigation models have been widely adopted by the web engineering community. HM3 is a navigation modelling method that we apply to the development of a system to support e-participatory budget formation. The system was first implemented without following a navigation modelling method. By comparing these two approaches, we detect navigational issues that could facilitate the users understanding of decision processes, and suggest guidelines to systematise the development of e-democracy applications.

Towards Knowledge Modeling for Sustainable Transport

The paradigm shift from the current energy consumption model towards a sustainable model requires to develop new behaviors and strategies. This is particularly relevant in domains like the public transport. Many providers are currently offering services to assist passengers to plan their routes. However, these approaches are often restricted to some specific area or transport medium. We suggest using a Linked Data perspective, which makes simpler to combine data from different sources, as well as extending and managing them. Moreover, it makes possible to enrich the basic model to the extent of developing a knowledge model, able to use semantic techniques to unfold even better strategies. In this paper we present a proposal in the transport domain, which refines a basic model into a Transmodel specification and later adds more information according to the IFOPT model. This defines a knowledge model, which can be used to develop sustainable transport strategies.

Modeling Open Accessibility Data of Public Transport

Nowadays linked open data (LOD) is a great challenge in the area of information technologies. In fact, some public administrations and organizations are working to open their data to citizens. LOD describes a method of publishing structured data to become more useful. Heterogeneous distributed data sources which have been published as LOD, could be integrated. The information of public transport networks is of public interest. A classic example is the route planning systems which combine data from different public transit networks.Besides, a relevant aspect of the public transit networks is the accessibility of this media which make possible the mobility of special needs people. Due to the large amount of these data, LOD provides the mechanism to publish them and to support new mobility services to the citizens. The aim of this work is present the process to define public transit data and their accessibility information as LOD from diverse data sources taking into account some of the main reference data models for public transport: IFOPT.