A decision support system for designing new services tailored to citizen profiles in a complex and distributed e-government scenario

https://doi.org/10.1016/j.datak.2008.06.005Get rights and content

Abstract

In this paper, we propose a new system aiming to support government agency decision makers to design new services tailored to citizen profiles in a complex and distributed e-government scenario. Specifically, our system assists government agency managers, who plan to activate new services, to identify those citizens who could gain the highest benefit from each of them. Managers can, then, exploit this information to decide what services should be activated and how they can be tailored to citizen needs and desires. Our system can handle more government agencies and a great number of citizens simultaneously; as a consequence, it is well suited for a complex e-government scenario. This paper first illustrates the proposed system; after this, it reports various experimental results; finally, it presents a comparison between our system and other related ones already presented in the literature.

Introduction

The term “e-government” is adopted to indicate the usage of Information and Communication Technologies for improving the interaction between citizens and government agencies [29]. For instance, e-government systems simplify citizen access to government services, allow government agencies to effectively deliver their services and enhance citizen participation in government and decision making activities.

In the last few years, both local and national governments are planning and implementing e-government programs worldwide [2], [38]. These initiatives have been generally welcomed by citizens, and several studies point out that many citizens and businesses require online services to their governments: for instance, in 2002, 68 million US citizens accessed government agency Web sites [29]. From the government agency point of view, there is a growing interest to deliver a large number of services through the Internet. For instance, presently, the average monthly number of accesses to Web pages relating to Hong Kong e-government system amounts to 230 millions; this system delivers over 1200 services [24]; in addition, the public expenditure of Hong Kong Government in Information and Communication Technologies grew from 910 millions (Hong Kong Dollars) in 1994 to 5236 millions in 2007 [23].

These considerations explain the enormous, both technological and scientific, efforts performed, in the last few years, for improving both the range and the quality of services delivered by government agencies online.

One of the most interesting research directions in this context consists of supporting the managers of government agencies to make their decisions about the definition of new services to be delivered to citizens [16], [18]. This support could be extremely relevant to an efficient management of available resources; in fact, many papers (e.g., [34], [42]) show that a suitable service planning policy may produce significant cost and time savings and, ultimately, may increase both the effectiveness and the productivity of Public Administration; in some cases (e.g., developing countries) it can compensate the lack of financial resources [6]. In addition, some papers (e.g., [42]) show that a key factor in the economic development of a community is represented by the activation of new services tailored to the profile of its members.

This paper aims to provide a contribution in this setting; in fact, it proposes a system for supporting government agency decision makers to design new services tailored to citizen needs and desires in a complex and distributed e-government scenario. Specifically, the proposed system aims to assist a government agency manager, who has a given budget for the activation of new services, to identify those citizens who could gain the highest benefit from each of them. Managers can exploit this information to decide what services should be activated and how they can be tailored to citizen needs and desires.

In order to properly manage citizen needs and desires, our system associates an appropriate profile [26] with each citizen accessing it. User profiles are suitable data structures that store demographic data, preferences, needs and past behaviours of the corresponding users. In the past, they have been extensively exploited in various application contexts, such as Web search [40], query answering [28], and e-commerce [14]. In our opinion, the knowledge of citizen profiles can improve both the effectiveness and the accuracy of an e-government system. Specifically, their usage enables the behaviour and needs of different citizens to be known and the proposals of an e-government system to be adapted to them. This can be extremely important for both citizens (who might receive only proposals related to services they really need) and government agencies (that might find those citizens who have the maximum benefit from a new service and/or might design new services that best match citizen expectations).

Interestingly enough, citizen profiles register also the past history of citizens and, therefore, they shed the light on the evolution of citizen needs on a temporal scale. This has a positive impact on decision-making activities; in fact, as stated in [42], unless there is a sudden crisis, the allocation of financial resources can be regarded as a (never ending) sequence of related decisions. As a consequence, the knowledge of the temporal evolution of citizen needs and desires is a key factor for the success of a planning policy.

As a further, interesting feature, in the context of e-government, citizen profiles are particularly rich and detailed since many interesting data about citizens are already owned by Public Administration offices. In our opinion, this fact is particularly important since the richness of user profiles is a key factor for software systems based on user modelling.

Our system is characterized by a distributed architecture that allows it to operate in a complex and distributed scenario. The distributed architecture adopted for our system is capable of preserving the autonomy of each government agency that can continue to use its own internal rules and work procedures to represent and handle its information. However, in our system, a citizen is provided with a virtual front-office that supplies a uniform interface for accessing services delivered by (independent) government agencies. As a consequence, a citizen has no need to know the internal organization of each agency and/or the relationships among agencies. In fact, these problems are faced by our system at a back-office level.

Interestingly enough, this choice is compliant with the ideas described in [15] and the suggestions outlined in [1], [27]. Specifically, [15] models Public Administration as an unstructured network of entities that exchange information objects each other in order to deliver citizen-oriented services; Ref. [1] illustrates a very simple paradigm, called one-stop e-government, that aims to integrate the whole spectrum of public services into a single framework.

In order to carry out its activity, our system handles two databases, namely: (i) a Service Database, that stores and manages information about services provided by government agencies; (ii) a Citizen Profile Database, that stores and handles information about citizens; most of this information (e.g., demographic data) is already owned by government agencies; the remaining part is derived and handled by monitoring citizens behaviour during their past accesses to e-government services.

In our system three different types of Handlers continuously operate and exchange information; these are: (i) Manager Handlers, that support managers of government agencies in their decision making activities; (ii) Citizen Handlers, that support citizens in their search of services of interest to them; (iii) Service Handlers, that handle service profiles and cooperate with Citizen Handlers to detect, among available services, those appearing the closest to citizen needs and profiles.

In order to carry out its activity, our system exploits two coefficients; the former states the benefits that a citizen would gain from the activation of a service; the latter denotes the cost that a government agency needs to pay to provide a certain citizen with a specific service. These coefficients are computed by exploiting the profiles of the citizens and the services into examination. They are used to define a suitable integer linear programming problem whose solution allows the Manager Handler to suggest the most promising services to activate in such a way as to comply with budget constraints and maximize citizen satisfaction.

The outline of this paper is as follows. Section 2 presents a detailed description of the proposed system; experiments carried out to test its performance are illustrated in Section 3. A detailed comparison between our system and other ones conceived to support decision making activities in e-government is presented in Section 4; finally, in Section 5, we draw our conclusions.

Section snippets

General overview

Before providing a description of our system we illustrate some real life scenarios which could benefit from it. In order to better illustrate these scenarios we observe that, in the last years and in many Western countries, Governments have been under pressure to design and deliver services in competition each other. In this case citizens can choose needed and/or desired services among multiple providers which can offer similar services under different conditions. In this scenario, the

Experimental results

This section illustrates the experiments that we have carried out to evaluate the performance of our system. Specifically, in Section 3.1 we provide some details about system implementation. In Section 3.2, we describe the datasets exploited for most of our experiments. Section 3.3 is devoted to measure the accuracy of our system. In Section 3.4, we analyze the impact of available budget on our system’s policies. Section 3.5 is devoted to verify the robustness of our system. The analysis of its

Related work

In the literature several systems devoted to support both government agencies and citizens in their decisions and choices have been proposed. In this section, we compare some of them with our own; before performing this comparison, we would like to point out that most of the existing systems are specific, i.e., they have been conceived to operate in specific contexts, such as health, public transports, and so on. On the contrary, our system is generic, i.e., it is capable of operating on all

Conclusions

In this paper, we have presented a system conceived to support managers of government agencies to design new services tailored to citizen profiles in a complex and distributed e-government scenario. Our system constructs and handles suitable profiles for involved citizens and services. These profiles are used to determine the benefits and the costs arising from the activation of a set of services. This information is used to define a 0/1 Integer Linear Programming problem whose solution

Pasquale De Meo is a Postdoctoral Researcher at the University Mediterranea of Reggio Calabria. He received his Laurea Degree in Electrical Engineering from the University Mediterranea of Reggio Calabria in May 2002 and his Ph.D. in System Engineering and Computer Science from the Dipartimento di Elettronica Informatica e Sistemistica, University of Calabria in February 2006. He was the winner of the AI*IA 2006 award for the Best Italian Ph.D. Thesis in Artificial Intelligence. His research

References (42)

  • D. Bertsimas et al.

    Introduction to Linear Optimization

    (1997)
  • P.K. Chan

    A non-invasive learning approach to building Web user profiles

  • S. Chaudhary et al.

    Architecture of sensor based agricultural information system for effective planning of farm activities

  • J. Clarke et al.

    Creating Citizen-Consumers: Changing Identities in the Remaking of Public Services

    (2007)
  • G.B. Dantzig, Maximization of linear function of variables subject to linear inequalities, in: T.C. Koopmans (Ed.),...
  • P. De Meo et al.

    An XML-based multi-agent system for supporting online recruitment services

    IEEE Transactions on Systems, Man ad Cybernetics – Part A

    (2007)
  • P. De Meo et al.

    Utilization of intelligent agents for supporting citizens in their access to e-government services

    Web Intelligence and Agent Systems Journal

    (2007)
  • M. Deshpande et al.

    Item-based Top-N recommendation algorithms

    ACM Transactions on Information Systems

    (2004)
  • G. Paiva Dias et al.

    A simple model and a distributed architecture for realizing one-stop e-government

    Electronic Commerce Research and Applications

    (2007)
  • S. Flessa

    Where efficiency saves lives: a linear programme for the optimal allocation of health care resources in developing countries

    Health Care Management Science

    (2000)
  • K. Goto, Y. Kambayashi, A new passenger support system for public transport using mobile database access, in:...
  • Cited by (22)

    • Recommender system application developments: A survey

      2015, Decision Support Systems
      Citation Excerpt :

      In this section, we will review the developments and applications of e-government recommender systems, in particular e-government Web interface personalization and adaptation and e-government service recommendation, which include government-to-citizen (G2C) and government-to-business (G2B) services. To support citizens in their access to personalized and adapted services supplied by public administration offices, a multi-agent system was presented by De Meo et al. [76]. The proposed system identifies and suggests the most interesting services for a user by considering both the user's profile and the profile of the device being used.

    • A standard language for service delivery: Enabling understanding among stakeholders

      2012, Computer Standards and Interfaces
      Citation Excerpt :

      User profiles are suitable data structures that store demographic data, preferences, needs and past behaviors of the corresponding users. A system proposed in [49] that is able to support the matching of supply and demand of public services consists of three main parts. Firstly, citizen handlers help citizens access e-government services.

    • Automatic Bill Recommendation for Statehouse Journalists

      2023, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    View all citing articles on Scopus

    Pasquale De Meo is a Postdoctoral Researcher at the University Mediterranea of Reggio Calabria. He received his Laurea Degree in Electrical Engineering from the University Mediterranea of Reggio Calabria in May 2002 and his Ph.D. in System Engineering and Computer Science from the Dipartimento di Elettronica Informatica e Sistemistica, University of Calabria in February 2006. He was the winner of the AI*IA 2006 award for the Best Italian Ph.D. Thesis in Artificial Intelligence. His research interests include multi-agent systems, user modeling, personalization, XML, Cooperative Information Systems, Folksonomies.

    Giovanni Quattrone is a Postdoctoral Researcher at the University Mediterranea of Reggio Calabria. He received his Laurea Degree in Electrical Engineering from the University Mediterranea of Reggio Calabria in July 2003 and his Ph.D. in Computer Science, Biomedical Engineering and Telecommunications Engineering from the University Mediterranea of Reggio Calabria in February 2007. His research interests include user modeling, intelligent agents, e-commerce, knowledge extraction and representation, scheme integration, XML, Cooperative Information Systems, Folksonomies.

    Domenico Ursino received the Laurea Degree in Computer Engineering from the University of Calabria in July 1995. He received his Ph.D. in System Engineering and Computer Science from the University of Calabria in January 2000. From October 2000 to January 2005 he was an Assistant Professor at University Mediterranea of Reggio Calabria. Currently he is an Associate Professor at the same University. His research interests include multi-agent systems, personalized and device-adaptive e-services, knowledge extraction and representation, scheme integration, semi-structured data and XML, Cooperative Information Systems, Folksonomies.

    View full text