Search

The security of critical infrastructures is of paramount importance nowadays due to the growing complexity of components and applications. This paper collects the contributions to the industry dissemination session within the 14th International Conference on Critical Information Infrastructures Security (CRITIS 2019). As such, it provides an overview of recent practical experience reports in the field of critical infrastructure protection (CIP), involving major industry players. The set of cases reported in this paper includes the usage of serious gaming for training infrastructure operators, integrated safety and security management in the chemical/process industry, risks related to the cyber-economy for energy suppliers, smart troubleshooting in the Internet of Things (IoT), as well as intrusion detection in power distribution Supervisory Control And Data Acquisition (SCADA). The session has been organized to stimulate an open scientific discussion about industry challenges, open issues and future opportunities in CIP research.

Marked mix-nets were introduced by Pereira and Rivest as a mechanism to allow very efficient mixing that ensures privacy but at the cost of not guaranteeing integrity. This is useful in a number of e-voting schemes such as STAR-Vote and Selene. However, the proposed marked mix-net construction comes with no proof of security and, as we show in this paper, does not provide privacy even in the presence of a single corrupt authority. Fortunately, the attack that we present is easy to prevent and we show several possible ways to address it. Finally while the original marked mix-net paper worked with ElGamal, we identify conditions that the adopted encryption scheme should satisfy in order to be appropriate for a marked mix-net. This opens the possibility of building marked mix-nets based on intractability assumptions which are believed to hold in the presence of a quantum computer.

Recent years have seen many privacy violations that have cost both the users of software systems and the businesses that run them in a variety of ways. One potential cause of these violations may be the ad hoc nature of the implementation of privacy measures within software systems, which may stem from the poor representation of privacy within many Software Development LifeCycle (SDLC) processes. We propose to give privacy a higher priority within the SDLC through the creation of a confederated Privacy-Aware SDLC (PASDLC) which incorporates the Data Protection Impact Assessment (DPIA) lifecycle. The PASDLC brings stakeholders of the software system closer together through the implementation of multiple interception points, whilst prompting the stakeholders to consider privacy within the software system. We consider many challenges to the creation of the PASDLC, including potential communication issues from confederating the processes of a SDLC and the effective measurement of privacy as an attribute of a software system. Postprint

Our life depends more and more on intelligent machines that help us carry out our work or have completely taken our place in tasks which used to be wholly human. This also applies to the field of care and assistance for the sick and fragile: indeed, carebots are increasingly present by the patient’s bed and can collaborate in care work. However carebots are not objects like others that we use in our everyday life: unlike the tools through which we do things, robots have (and will more and more have) some level of autonomy. We intend to consider the type of relationship it is right to build with these devices and ask whether intelligent robots deserve some moral and legal relevance. The robot may be seen as a slave or an entity with which we can form friendships or loving relationships: we will maintain that it is our responsibility to prepare for future scenarios in which increasingly intelligent, autonomous machines will be not mere tools, but significant life companions.

In over fve years, Bornmann, Stefaner, de Moya Anegon, and Mutz (2014b) and Bornmann, Stefaner, de Moya Anegón, and Mutz (2014c, 2015) have published several releases of the www.excellencemapping.net tool revealing (clusters of) excellent institutions worldwide based on citation data. With the new release, a completely revised tool has been published. It is not only based on citation data (bibliometrics), but also Mendeley data (altmetrics). Thus, the institutional impact measurement of the tool has been expanded by focusing on additional status groups besides researchers such as students and librarians. Furthermore, the visualization of the data has been completely updated by improving the operability for the user and including new features such as institutional profle pages. In this paper, we describe the datasets for the current excellencemapping.net tool and the indicators applied. Furthermore, the underlying statistics for the tool and the use of the web application are explained.

This paper presents a demonstration of our PAAMS 2021 paper using data-driven analysis of airport terminal operations and An Agent-based Airport Terminal Operations Model Simulator (AATOM). The goal of this paper is to demonstrate and analyze the impact of the current COVID-19 and future pandemic-related measures on airport terminal operations and to identify plans that airport management agents can take into account to control the flow of passengers in a safe, efficient, secure and resilient way. To analyze the impact of the identified COVID-19 measures on the airport operations, the existing agent-based AATOM model was need to be modified in order to implement these measures. In this paper, we illustrate a demo of a developed simulator tool by investigating the effects of different degrees of physical distancing rules among agents on the performances of the airport. In the demo session the attendees will have the possibility to (i) work with the simulator tool on different relevant parameters regarding different sections and agents in the airport; (ii) view and analyze different performance indicator analyzers of the simulator.

The ever-increasing number of published scientific articles has prompted the need for automated, data-driven approaches to summarizing the content of scientific articles. The Computational Linguistics Scientific Document Summarization Shared Task (CL-SciSumm 2019) has recently fostered the study and development of new text mining and machine learning solutions to the summarization problem customized to the academic domain. In CL-SciSumm, a Reference Paper (RP) is associated with a set of Citing Papers (CPs), all containing citations to the RP. In each CP, the text spans (i.e., citances) have been identified that pertain to a particular citation to the RP. The task of identifying the spans of text in the RP that most accurately reflect the citance is addressed using supervised approaches. This paper proposes a new, more effective solution to the CL-SciSumm discourse facet classification task, which entails identifying for each cited text span what facet of the paper it belongs to from a predefined set of facets. It proposes also to extend the set of traditional CL-SciSumm tasks with a new one, namely the discourse facet summarization task. The idea behind is to extract facet-specific descriptions of each RP consisting of a fixed-length collection of RP’s text spans. To tackle both the standard and the new tasks, we propose machine learning supported solutions based on the extraction of a selection of discriminating words, called pivot words. Predictive features based on pivot words are shown to be of great importance to rate the pertinence and relevance of a text span to a given facet. The newly proposed facet classification method performs significantly better than the best performing CL-SciSumm 2019 participant (i.e., the classification accuracy has increased by + 8%), whereas regression methods achieved promising results for the newly proposed summarization task.

In this demo paper, we describe a Multi Agent System (MAS) application for monitoring and evaluation of the ethical behavior of customer service agents in an online chat point w.r.t their institution/company’s codes of ethics and conduct. This MAS is realized using the state-of-the-art JaCaMo framework, and can be practically demonstrated to an audience.

Smart Study is an educational platform allowing learners to continue using pen and paper for learning purposes. The platform allow children to solve exercises in the classical way, using a pen and paper. Afterwards, the handwritten solutions are digitalized, automatically corrected and animated feedback is provided by means of a tablet. By using pen and paper, readwrite learners can still use their preferred learning method while for all other learners it allows them to practice handwriting. The use of a tablet for feedback and for providing information makes it a lightweight and mobile platform, easy to use by children. The platform was evaluated with 15 children age 10 to 11. This evaluation showed that the children were more motivated than usual and enjoyed working with the platform. Also their teachers were enthusiastic but questioned the cost for its use in a classroom environment. However, the platform could also perform well in a remediation situation, home or hospital environment.

International audience; We propose a compositional specification theory for reason- ing about components that interact by synchronisation of input and out- put (I/O) actions, in which the specification of a component constrains the temporal ordering of interactions with the environment. Such a the- ory is motivated by the need to support composability of components, in addition to modelling environmental assumptions, and reasoning about run-time behaviour. Models can be specified operationally by means of I/O labelled transition systems augmented by an inconsistency predi- cate on states, or in a purely declarative manner by means of traces. We introduce a refinement preorder that supports safe-substitutivity of components. Our specification theory includes the operations of parallel composition for composing components at run-time, logical conjunction for independent development, and quotient for incremental development. We prove congruence properties of the operations and show correspon- dence between the operational and declarative frameworks.

Logical frameworks like LF allow to specify the syntax and (natural deduction) inference rules for syntax/proof-checking a wide variety of logical systems. A crucial feature that is missing for prototyping logics is a way to specify basic proof automation. We try to alleviate this problem by generating \(\lambda \)Prolog (ELPI) inference predicates from logic specifications and controlling them by logic-independent helper predicates that encapsulate the prover characteristics. We show the feasibility of the approach with three experiments: We directly automate ND calculi, we generate tableau theorem provers and model generators.

The Semantic Web: ESWC 2019 Satellite Events - ESWC 2019 Satellite Events, Portorož, Slovenia, June 2-6, 2019, Revised Selected Papers

This paper provides an overview of the main developments of the Tech United Eindhoven RoboCup @Home team. Tech United uses an advanced world modeling system called the Environment Descriptor. It allows straightforward implementation of localization, navigation, exploration, object detection & recognition, object manipulation and robot-robot cooperation skills based on the most recent state of the world. Other important features include object and people detection via deep learning methods, a GUI, speech recognition, natural language interpretation and a chat interface combined with a conversation engine. Recent developments that aided with obtaining the victory during RoboCup 2019 include pointing detection, usage of HSR’s display, a people detector and the addition of a custom keyboard in the chat interface.

To make the geometrical cognitive process more interactive, we produced teaching aids (tangible models, graphic tablets) that help students in visualizing their geometrical-analytical investigations of the architectural artifacts and enhance their spatial prefiguration and critical form-reading skills, three-dimensional thinking and geometrical reading of shapes. Then, we looked for a medium suitable to create simple three-dimensional models, not only observable, like virtual models, not only tangible, like physical models proposed in the design studios, but also dynamic, using multiple media and languages in the same training message. As an example, we present here an interdisciplinary lesson between Calculus and Architectural Drawing and Survey Laboratory about developable surfaces, experimented on first year students of the bachelor program in Architecture. The lesson is based on the use of a graphic tablet and some origami inspired models: it summarizes the geometric description of a pyramid and a cloister vault of equal height and equal orthographic projection on the horizontal plane. We saw that tackling the same topic in both teaching contexts is not a useless overlap, but a stimulus to compare different languages and methods. 2D and 3D paper models of artifacts—and of projective reduction from 3D to the plane—aid spatial intuition and the subtle exercise of controlling mental images which replace artifacts, turning 3D configurations into signifying images. Moreover, this experience stimulates reading and evaluation of the drawn geometry (ruled surfaces, projections, developments), increasing critical sense in reading the built environment.

Deductive software verification aims at formally verifying that all possible behaviors of a given program satisfy formally defined, possibly complex properties, where the verification process is based on logical inference. We follow the trajectory of the field from its inception in the late 1960s via its current state to its promises for the future, from pen-and-paper proofs for programs written in small, idealized languages to highly automated proofs of complex library or system code written in mainstream languages. We take stock of the state-of-art and give a list of the most important challenges for the further development of the field of deductive software verification.

As research processes change due to technological developments in how data is collected, stored and used, so must consent methods. Dynamic consent is an online mechanism allowing research participants to revisit consent decisions they have made about how their data is used. Emerging from bio-banking where research data is derived from biological samples, dynamic consent has been designed to address problems with participant engagement and oversight. Through discussion that emerged during a workshop run at the IFIP 2018 Summer School, this paper explores wider research problems could be addressed by dynamic consent. Emergent themes of research design, expectation management and trust suggested overarching research problems which could be addressed with a longer term view of how research data is used, even if that use is unknown at the point of collection. We posit that the existing model of dynamic consent offers a practical research approach outside of bio-banking. Published version

Reproducibility of Science is considered as one of the main principles of the scientific method, and refers to the ability of an experiment to be accurately reproduced, by third person, in complex experiment every detail matters to ensure the correct reproducibility. In the context of the ICCS 2011, Elsevier organized the executable paper grand challenge a contest to improve the way scientific information is communicated and used. While during this contest the focus was on developing methods and technique to realize the idea of executable papers, in this paper we focus on the operational issues related to the creation a viable service with a predefined QoS.

Distributed Ledgers (DLs), also known as blockchains, provide decentralised, tamper-free registries of transactions among partners that distrust each other. For the scientific community, DLs have been proposed to decentralise and make more transparent each step of the scientific workflow. For the particular case of dissemination and peer-reviewing, DLs can provide the cornerstone to realise open decentralised publishing systems where social interactions between peers are tamper-free, enabling trustworthy computation of bibliometrics. In this paper, we propose the use of DL-backed smart contracts to track a subset of social interactions for scholarly publications in a decentralised and reliable way, yielding Smart Papers. We show how our Smart Papers approach complements current models for decentralised publishing, and analyse cost implications.

Part 1: Long and Short Papers (Continued); International audience; In this paper we introduce TicQR – a photo-based checkbox-enabled interface which bridges the physical and digital document domains, allowing automatic download or processing of useful data from paper documents. There is a long demonstrated need for people to be able to connect between printed material and digital information and services. By using a combination of image recognition and QR codes we are able to detect user marks on paper documents via a single photograph taken with a standard smart phone. This information can then be used to access the equivalent digital content, save contacts or URLs, or even order goods directly from local retailers.

In digital libraries, much of the reading activity is still done on printed copies of documents. We show how digital pen and paper technologies can be used to support readers by automatically creating interactive paper versions of digital documents during the printing process that enable users to activate embedded hyperlinks to other documents and services from printed versions. The approach uses a special printer driver that allows information about hyperlinks to be extracted and stored at print time. Users can then activate hyperlinks in the printed document with a digital pen.

According to the data from the Scopus publication database, as analyzed in several recent studies, more than 70,000 papers have been published in the area of Software Engineering (SE) since late 1960’s. According to our recent work, 43% of those papers have received no citations at all. Since citations are the most commonly used metric for measuring research (academic) impact, these figures raise questions (doubts) about the (non-existing) impact of such a large set of papers. It is a reality that typical academic reward systems encourage researchers to publish more papers and do not place a major emphasis on research impact. To shed light on the issue of volume (quantity) versus citation-based impact of SE research papers, we conduct and report in this paper a quantitative bibliometrics assessment in four aspects: (1) quantity versus impact of different paper types (e.g., conference versus journal papers), (2) ratios of uncited (non-impactful) papers, (3) quantity versus impact of papers originating from different countries, and (4) quantity versus impact of papers by each of the top-10 authors (in terms of number of papers). To achieve the above objective, we conducted a quantitative exploratory bibliometrics assessment, comprised of four research questions, to assess quantity versus impact of SE papers with respect to the aspects discussed above. We extracted the data through a systematic, automated and repeatable process from the Scopus paper database, which we also used in two previous papers. Our results show that the distribution of SE publications has a major inequality in terms of impact overall, and also when categorized in terms of the above four aspects. The situation in the SE literature is similar to the other areas of science as studied by previous bibliometrics studies. Also, among our results is the fact that journal articles and conference papers have been cited 12.6 and 3.6 times on average, confirming the expectation that journal articles have more impact, in general, than conference papers. Also, papers originated from English-speaking countries have in general more visibility and impact (and consequently citations) when compared to papers originated from non-English-speaking countries. Our results have implications for improvement of academic reward systems, which nowadays mainly encourage researchers to publish more papers and usually neglect research impact. Also, our results can help researchers in non-English-speaking countries to consider improvements to increase their research impact of their upcoming papers., Vahid Garousi was partially supported by several internal grants provided by the Hacettepe University and the Scientific and Technological Research Council of Turkey (TUBITAK). Joao M. Fernandes was supported by FCT - Fundacao para a Ciencia e Tecnologia within the Project Scope UID/CEC/00319/2013.

This book constitutes the refereed proceedings of the 5th Computer Games Workshop, CGW 2016, and the 5th Workshop on General Intelligence in Game-Playing Agents, GIGA 2016, held in conjunction with the 25th International Conference on Artificial Intelligence, IJCAI 2016, in New York, USA, in July 2016.The 12 revised full papers presented were carefully reviewed and selected from 25 submissions. The papers address all aspects of artificial intelligence and computer game playing. They discuss topics such as Monte-Carlo methods; heuristic search; board games; card games; video games; perfect and imperfect information games; puzzles and single player games; multi-player games; combinatorial game theory; applications; computational creativity; computational game theory; evaluation and analysis; game design; knowledge representation; machine learning; multi-agent systems; opponent modeling; planning.

While collaborative filtering and citation analysis have been well studied for research paper recommender systems, content-based approaches typically restrict themselves to straightforward application of the vector space model. However, various types of metadata containing potentially useful information are usually available as well. Our work explores several methods to exploit this information in combination with different similarity measures.

In this paper we present a survey of the work on type-based useless-code elimination for higher-order functional programs. After some historical remarks on the motivations and early approaches we give an informal but complete account of the techniques and results developed at the Computer Science Department of the University of Torino. In particular, we focus on the fact that, for each of the type-based techniques developed, there is an optimal program simplification.

Paper documents are often described as ‘information rich’, in contrast with electronic documents. This ethnographic study examines Lotus NOTES in a sub-section of the Irish civil service, with particular reference to the concurrent use of electronic and paper documents. The sub-section examines disagreements with regard to claims by Irish citizens for particular government benefits. The study describes how meta-information contained in paper case files is perceived as necessary for the work of the organization, thus restricting the use of electronic case files in NOTES as a shared information system. However, this reliance on paper files derives not only from the information rich properties of paper documents, but also from the desire of some workers to protect their occupational status by defining, as necessary for their job, information which is only available in paper documents and which only they can interpret. This dependence on paper documents also reduces the amount of information that can be shared within the organization. This paper suggests that, only if the perceived threat posed by the information system were reduced in some way would user innovations in work practices and greater sharing of information within the organization become possible.

We present a demonstration system for incremental spoken human–machine dialogue for task-centric domains that includes a controller for verbal and nonverbal behavior for virtual agents. The dialogue management components can handle uncertainty in input and resolve it interactively with high responsivity, and state tracking is aware of momentary events such as interruptions by the user. Aside from adaptable dialogue strategies, such as for grounding, the system includes a multimodal floor management controller that attempts to limit the influence of idiosyncratic dialogue behavior on the part of our primary user groups – older adults and people with cognitive impairments – both of which have previously participated in pilot studies using the platform.

This book constitutes the thoroughly refereed post-conference proceedings of the 10th International Conference on Adaptive Multimedia Retrieval, AMR 2012, held in Copenhagen, Denmark, in October 2012.The 17 revised full papers presented were carefully reviewed and selected from numerous submissions. The papers cover topics of state of the art contributions, features and classification, location context, language and semantics, music retrieval, and adaption and HCI.

In this paper, the influence and improvement of the localization accuracy achieved using a fingerprint database with information coming from different channels and radio signal strength levels is evaluated. This study uses IEEE 802.15.4 networks with different power levels and carrier frequency channels in the 2.4 GHz band. Experimental results show that selecting part of this information with a cleverer data processing can provide similar or better localization accuracy than using the whole database.

International audience; This book constitutes the refereed proceedings of the Computer Games Workshop, CGW 2013, held in Beijing, China, in August 2013, in conjunction with the Twenty-third International Conference on Artificial Intelligence, IJCAI 2013. The 9 revised full papers presented were carefully reviewed and selected from 15 submissions. The papers cover a wide range of topics related to computer games. They discuss six games that are played by humans in practice: Chess, Domineering, Chinese Checkers, Go, Goofspiel, and Tzaar. Moreover, there are papers about the Sliding Tile Puzzle, an application, namely, Cooperative Path-Finding Problems, and on general game playing.

We study a population involved in a cyclic game of three strategies – the rock-paper-scissors game – whose agents interact through groups of three individuals (triplets), considering the possibility that two weak agents cooperate and beat a strong one. In a wide range of parameters the system presents a stable heteroclinic cycle, which implies that in a finite population some of the strategies become extinct and others survive. We find that the cooperation within triplets only benefits the survival of the strategy if the cooperation probability is above a certain threshold. We study the survival probabilities of the different strategies as a function of the cooperation parameters through a analytic approximation and compare with simulations, obtaining a good agreement. Results are generalizable to other systems with heteroclinic cycles. Fil: Hernández Lahme, Damián Gabriel. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Comisión Nacional de Energía Atómica. Centro Atómico Bariloche; Argentina Fil: Zanette, Damian Horacio. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Comisión Nacional de Energía Atómica. Centro Atómico Bariloche; Argentina

Dexter was my manager during the first 8 months or so of my one year stay at the IBM Research Center at Yorktown Heights in the mid eighties. Our interaction in terms of scientific output was very slim -- it consisted of just a single paper, titled "Limits for automatic verification of finite-state concurrent systems." The paper appeared in 1986 in Information Processing Letters and had just 3 pages. Still, it has 275 citations on Google scholar, so more than 90 citations per page. I believe it is the best per page citation ratio for both of us.

A collection of coauthored papers is the new norm for doctoral dissertations in the natural and biomedical sciences, yet there is no consensus on how to partition authorship credit between PhD candidates and their coauthors. Guidelines for PhD programs vary but tend to specify only a suggested range for the number of papers to be submitted for evaluation, sometimes supplemented with a requirement for the PhD candidate to be the principal author on the majority of submitted papers. Here I use harmonic counting to quantify the actual amount of authorship credit attributable to individual PhD graduates from two Scandinavian universities in 2008. Harmonic counting corrects for the inherent inflationary and equalizing biases of routine counting methods, thereby allowing the bibliometrically identifiable amount of authorship credit in approved dissertations to be analyzed with unprecedented accuracy. Unbiased partitioning of authorship credit between graduates and their coauthors provides a post hoc bibliometric measure of current PhD requirements, and sets a de facto baseline for the requisite scientific productivity of these contemporary PhD’s at a median value of approximately 1.6 undivided papers per dissertation. Comparison with previous census data suggests that the baseline has shifted over the past two decades as a result of a decrease in the number of submitted papers per candidate and an increase in the number of coauthors per paper. A simple solution to this shifting baseline syndrome would be to benchmark the amount of unbiased authorship credit deemed necessary for successful completion of a specific PhD program, and then monitor for departures from this level over time. Harmonic partitioning of authorship credit also facilitates cross-disciplinary and inter-institutional analysis of the scientific output from different PhD programs. Juxtaposing bibliometric benchmarks with current baselines may thus assist the development of harmonized guidelines and transparent transnational quality assurance procedures for doctoral programs by providing a robust and meaningful standard for further exploration of the causes of intra- and inter-institutional variation in the amount of unbiased authorship credit per dissertation. Paid Open Access

Organizational best practices are unstructured, emergent processes that freely coordinate actors engaged in reaching organizations' goals. In recent years we are witnessing the wide adoption of social software (blogs, microblogs, wiki, forums, shared calendars, etc.) as primary technological tools to support organizational best practices, fostering their creation, evolution and sharing, allowing their continuous refinement and alignment with the organization's mission and evolving know-how. While organizational best practices and social software tools are good candidates to support specific processes within the organization (and among organizations) they also present several issues, when compared to classic BPM tools - those based on structured coordination and well-defined process models: since they have no explicit representation it is hard to analyze them (by analytic techniques or by simulation), to monitor their evolution and to support their execution; moreover it is hard to extract explicit knowledge from them. In this paper we present a set of tools that complement social software in creating a real coordination platform, mitigating some of the aforementioned issues.

Information technology has great potential for transforming the health care system, improving quality of care. With the increasing expansion of health information systems, the Electronic Health Record (EHR) has become one of the finest sources for clinical information aggregators in the context of digital health. The EHR is a core part of a hospital information system, as well as a service on duty of the patient to improve the treatment of patients. It can be considered as a longitudinal electronic record of patient heath information, for example vital signs, medical history or laboratory data, generated by one or more encounters in any care delivery setting. As the EHR offers many potential opportunities for healthcare systems, it is important to take steps to improve the system. With this in mind, a study of the features present on a Portuguese EHR was made. The basis of this study was an adoption model that evaluates the EHR system accordingly to its current features. After this study, the EHR will be ranked into one of the existing eight stages.

Critical infrastructure (CI) services are consumed by the society constantly and we expect them to be available 24 hours a day. CIs can be mutually dependent on each other and a failure in one infrastructure can cascade to another interdependent infrastructure to cause service disruptions. Methods to better assess and monitor CIs and their interdependencies in order to predict possible risks have to be developed. In this work, we present a method for CI analysis to identify critical entities in CIs at a management/organisational level as well as at a technical level supported by the PROTOS-MATINE model for dependency analysis.

We demonstrate a collaborative knowledge management platform in which communities representing autonomously developed information systems build ontologies to achieve semantic interoperabil- ity between those systems. The tool is called GOSPL, which stands for Grounding Ontologies with Social Processes and natural Language, and supports the method bearing the same name. Ontologies in GOSPL are hybrid, meaning that concepts are both described informally in natural language and formally. Agreements on these two levels are made simulta- neously and the social interaction between and across communities drive the ontology evolution process

Security and trust policy models.- security protocol design and analysis.- formal models of trust and reputation.- logics for security and trust.- distributed trust management systems.- trust-based reasoning.- digital assets protection.- data protection.- privacy and ID issues.- information flow analysis.- language-based security.- security and trust aspects of ubiquitous computing.- validation/analysis tools.- web service security/trust/privacy.- grid security.- security risk assessment.- case studies.

In this paper we analyze the use of different types of statistical tests for the correlation of anomaly detection alerts. We show that the Granger Causality Test, one of the few proposals that can be extended to the anomaly detection domain, strongly depends on good choices of a parameter which proves to be both sensitive and difficult to estimate. We propose a different approach based on a set of simpler statistical tests, and we prove that our criteria work well on a simplified correlation task, without requiring complex configuration parameters.

We experiment a new influence function in our information retrieval method that uses the degree of fuzzy proximity of key terms in a document to compute the relevance of the document to the query. The model is based on the idea that the closer the query terms in a document are to each other the more relevant the document. Our model handles Boolean queries but, contrary to the traditional extensions of the basic Boolean information retrieval model, does not use a proximity operator explicitly. A single parameter makes it possible to control the proximity degree required. To improve our system we use a stemming algorithm before indexing, we take a specific influence function and we merge fuzzy proximity result lists built with different width of influence function. We explain how we construct the queries and report the results of our experiments in the ad-hoc monolingual French task of the CLEF 2006 evaluation campaign.