Today, handheld devices can accommodate a large amount of different resources. Thus, a considerable effort is often required to mobile users in order to search for the resources suitable for the specific circumstance. Further, this effort rarely brings to a satisfactory result. To ease this work, resource recommenders have been proposed in the last years. Typically, the recommendation is based on recognizing the current situations of the users and suggesting them the appropriate resources for those situations. The recognition task is performed by exploiting contextual information and preferably without using any explicit input from the user. To this aim, we propose to adopt a collaborative scheme based on an emergent paradigm. The underlying idea is that simple individual actions can lead to an emergent collective behavior that represents an implicit form of contextual information. We show how this behavior can be extracted by using a multi-agent scheme, where agents do not directly communicate amongst themselves, but rather through the environment. The multi-agent scheme is structured into three levels of information processing. The first level is based on a stigmergic paradigm, in which marking agents leave marks in the environment in correspondence to the position of the user. The accumulation of such marks enables the second level, a fuzzy information granulation process, in which relevant events can emerge and are captured by means of event agents. Finally, in the third level, a fuzzy inference process, managed by situation agents, deduces the user situations from the underlying events. The proposed scheme is evaluated on a set of representative real scenarios related to meeting events. In all the scenarios, the collaborative situation-aware scheme promptly recognizes the correct situations, except for one case, thus proving its effectiveness

Situation awareness is a computing paradigm which allows applications to sense parameters in the environment, comprehend their meaning and project their status in the next future. In collaborative situation awareness, a challenging area in the field of Ambient Intelligence applications, situation patterns emerge from users collective behavior. In this paper we introduce a multi-agent system that exploits positioning information coming from mobile devices to detect the occurrence of user0 s situations related to social events. In the functional view of the system, the first level of information processing is managed by marking agents which leave marks in the environment in correspondence to the users0 positions. The accumulation of marks enables a stigmergic cooperation mechanism, generating short-term memory structures in the local environment. Information provided by such structures is granulated by event agents which associate a certainty degree with each event. Finally, an inference level, managed by situation agents, deduces user situations from the underlying events by exploiting fuzzy rules whose parameters are generated automatically by a neuro-fuzzy approach. Fuzziness allows the system to cope with the uncertainty of the events. In the architectural view of the system, we adopt semantic web standards to guarantee structural interoperability in an open application environment. The system has been tested on different real-world scenarios to show the effectiveness of the proposed approach.

DC* (Double Clustering with A*) is an algorithm capable of generating highly interpretable fuzzy information granules from preclassified data. These information granules can be used as bulding-blocks for fuzzy rule-based classifiers that exhibit a good tradeoff between interpretability and accuracy. DC* relies on A* for the granulation process, whose efficiency is tightly related to the heuristic function used for estimating the costs of candidate solutions. In this paper we propose a new heuristic function that is capable of exploiting class information to overcome the heuristic function originally used in DC* in terms of efficiency. The experimental results show that the proposed heuristic function allows huge savings in terms of computational effort, thus making DC* a competitive choice for designing interpretable fuzzy rule-based classifiers.

When approaching real-world problems with intelligent systems, an interaction with user is often expected. However, data-driven models are usually evaluated only in terms of accuracy, thus not involving users. In literature several works have been proposed for defining measures for interpretability assessment, however, such measures are mostly based on a structural evaluation. For this reason, we investigated a new methodology for assessing interpretability based on semantic cointension. The objective of this work is to provide empirical evidence about the usefulness of semantic cointension in facing a medical problem, namely the prediction of prognosis in Immunoglobulin A Nephropathy. An experimental session has been conducted, where fuzzy rule-based classifiers have been modeled, which are highly interpretable from the structural viewpoint. Results show that through the notion of semantic cointension it is possible to perform a semantic-driven assessment of interpretability, which also takes into account the overall fuzzy inference schema.

In this paper we compare two algorithms that are capable of generating fuzzy partitions from data so as to verify a number of interpretability constraints: Hierarchical Fuzzy Partitioning (HFP) and Double Clustering with A* (DC*). Both algorithms exhibit the distinguishing feature of self-determining the number of fuzzy sets in each fuzzy partition, thus relieving the user from the selection of the best granularity level for each input feature. However, the two algorithms adopt very different approaches in generating fuzzy partitions, thus motivating an extensive experimentation to highlight points of strength and weakness of both. The experimental results show that, while HFP is on the average more efficient, DC* is capable of generating fuzzy partitions with a better trade-off between interpretability and accuracy, and generally offers greater stability with respect to its hyper-parameters.

The common practices of machine learning appear to be frustrated by a number of theoretical results denying the possibility of any meaningful implementation of a “superior” learning algorithm. However, there exist some general assumptions that, even when overlooked, preside the activity of researchers and practitioners. A thorough reflection over such essential premises brings forward the meta-learning approach as the most suitable for escaping the long-dated riddle of induction claiming also an epistemologic soundness. Several examples of meta-learning models can be found in literature, yet the combination of computational intelligence techniques with meta-learning models still remains scarcely explored. Our contribution to this particular research line consists in the realisation of Mindful, a meta-learning system based on the neuro-fuzzy hybridisation. We present the Mindful system firstly situating it inside the general context of the meta-learning frameworks proposed in literature. Finally, a complete session of experiments is illustrated, comprising both base-level and meta-level learning activity. The appreciable experimental results underline the suitability of the Mindful system for managing past accumulated learning experience while facing novel tasks.

In computing with words (CWW), knowledge is linguistically represented and has an explicit semantics defined through fuzzy information granules. The linguistic representation, in turn, naturally bears an implicit semantics that belongs to users reading the knowledge base; hence a necessary condition for achieving interpretability requires that implicit and explicit semantics are cointensive. Interpretability is definitely stringent when knowledge must be acquired from data through inductive learning. Therefore, in this paper we propose a methodology for designing interpretable fuzzy models through semantic cointension. We focus our analysis on fuzzy rule-based classifiers (FRBCs), where we observe that rules resemble logical propositions, thus semantic cointension can be partially regarded as the fulfillment of the "logical view", i.e. the set of basic logical laws that are required in any logical system. The proposed approach is grounded on the employment of a couple of tools: DCf, which extracts interpretable classification rules from data, and Espresso, that is capable of fast minimization of Boolean propositions. Our research demonstrates that it is possible to design models that exhibit good classification accuracy combined with high interpretability in the sense of semantic cointension. Also, structural parameters that quantify model complexity show that the derived models are also simple enough to be read and understood. © 2011 Elsevier Inc. All rights reserved.

The adoption of triangular fuzzy sets to define Strong Fuzzy Partitions (SFPs) is a common practice in the research community: due to their inherent simplicity, triangular fuzzy sets can be easily derived from data by applying suitable clustering algorithms. However, the choice of triangular fuzzy sets may be limiting for the modeling process. In this paper we focus on SFPs built up starting from cuts (points of separation between cluster projections on data dimensions), showing that a SFP based on cuts can always be defined by trapezoidal fuzzy sets. Different mechanisms to derive SFPs from cuts are presented and compared by employing DC*, an algorithm for extracting fuzzy information granules from classified data.

Granular computing is a problem solving paradigm based on information granules, which are conceptual entities derived through a granulation process. Solving a complex problem, via a granular computing approach, means splitting the problem into information granules and handling each granule as a whole. This leads to a multi-level view of information granulation, which permeates human reasoning and has a significant impact in any field involving both human-oriented and machine- oriented problem solving. In this chapter we examine a view of granular computing as a paradigm of human-inspired problem solving and information processing with multiple levels of granularity, with special focus on fuzzy information granulation. To support the importance of granulation with multiple levels, we present a multi-level approach for extracting well-defined and semantically sound fuzzy information granules from numerical data.

E-Government is becoming more attentive towards providing personalized services to citizens so that they can benefit from better services with less time and effort. To develop citizen-centered services, a fundamental activity consists in mining needs and preferences of users by identifying homogeneous groups of users, also known as user segments, sharing similar characteristics. Since the same user often has characteristics shared by several segments, in this work we propose an approach based on fuzzy clustering for inferring user segments that could be properly exploited to offer personalized services that better satisfy user needs and their expectations. User segments are inferred starting from data, gathered by questionnaires, which essentially describe demographic characteristics of users. For each derived segment a user profile is defined which summarizes characteristics shared by users belonging to that segment. Results obtained on a case study are reported in the last part of the paper.

In this work we propose an approach based on shape clustering for image retrieval. Firstly, shapes of objects contained into images are represented by means of Fourier descriptors. Then, a fuzzy clustering process is applied to automatically discover a set of shape prototypes representative of a number of semantic categories. The adopted fuzzy clustering algorithm is equipped with a mechanism of partial supervision that enables identification of shape categories by taking advantage of some domain knowledge expressed in terms of a set of labeled shapes. Successively, the derived shape prototypes are exploited in order to retrieve shapes similar to a shape query submitted by a user. The suitability of the proposed approach is shown through an experimental comparison on a benchmark dataset in terms of retrieval accuracy.

Computing with words (CWW) relies on linguistic representation of knowledge that is processed by operating at the semantical level defined through fuzzy sets. Linguistic representation of knowledge is a major issue when fuzzy rule based models are acquired from data by some form of empirical learning. Indeed, these models are often requested to exhibit interpretability, which is normally evaluated in terms of structural features, such as rule complexity, properties on fuzzy sets and partitions. In this paper we propose a different approach for evaluating interpretability that is based on the notion of cointension. The interpretability of a fuzzy rule-based model is measured in terms of cointension degree between the explicit semantics, defined by the formal parameter settings of the model, and the implicit semantics conveyed to the reader by the linguistic representation of knowledge. Implicit semantics calls for a representation of user's knowledge which is difficult to externalise. Nevertheless, we identify a set of properties - which we call "logical view" - that is expected to hold in the implicit semantics and is used in our approach to evaluate the cointension between explicit and implicit semantics. In practice, a new fuzzy rule base is obtained by minimising the fuzzy rule base through logical properties. Semantic comparison is made by evaluating the performances of the two rule bases, which are supposed to be similar when the two semantics are almost equivalent. If this is the case, we deduce that the logical view is applicable to the model, which can be tagged as interpretable from the cointension viewpoint. These ideas are then used to define a strategy for assessing interpretability of fuzzy rule-based classifiers (FRBCs). The strategy has been evaluated on a set of pre-existent FRBCs, acquired by different learning processes from a well-known benchmark dataset. Our analysis highlighted that some of them are not cointensive with user's knowledge, hence their linguistic representation is not appropriate, even though they can be tagged as interpretable from a structural point of view.

We propose an approach to integrate the KEEL software tool for knowledge discovery within the KNIME Analytics platform. The integration approach is non-invasive as it does not require the modification of source code in neither of the tools. As a result of the integration, it is possible to use the algorithms provided with KEEL — including many fuzzy methods — directly in KNIME workflows, thus taking the advantages of both tools. We report two simple integration examples, which show the effectiveness of the proposed approach in building data analysis workflows involving KEEL methods, possibly along with methods provided by other knowledge discovery tools like WEKA.

Recommender systems are systems capable of assisting users by quickly providing them with relevant resources according to their interests or preferences. The efficacy of a recommender system is strictly connected with the possibility of creating meaningful user profiles, including information about user preferences, interests, goals, usage data and interactive behavior. In particular, analysis of user preferences is important to predict user behaviors and make appropriate recommendations. In this paper, we present a fuzzy framework to represent, learn and update user profiles. The representation of a user profile is based on a structured model of user cognitive states, including a competence profile, a preference profile and an acquaintance profile. The strategy for deriving and updating profiles is to record the sequence of accessed resources by each user, and to update preference profiles accordingly, so as to suggest similar resources at next user accesses. The adaption of the preference profile is performed continuously, but in earlier stages it is more sensitive to updates (plastic phase) while in later stages it is less sensitive (stable phase) to allow resource recommendation. Simulation results are reported to show the effectiveness of the proposed approach.

In the era of the Web, there is urgent need for developing systems able to personalize the online experience of Web users on the basis of their needs. Web recommendation is a promising technology that attempts to predict the interests of Web users, by providing them with information and/or services that they need without explicitly asking for them. In this paper we propose NEWER, a usage-based Web recommendation system that exploits the potential of Computational Intelligence techniques to dynamically suggest interesting pages to users according to their preferences. NEWER employs a neuro-fuzzy approach in order to determine categories of users sharing similar interests and to discover a recommendation model as a set of fuzzy rules expressing the associations between user categories and relevances of pages. The discovered model is used by a online recommendation module to determine the list of links judged relevant for users. The results obtained on both synthetic and real-world data show that NEWER is effective for recommendation, leading to a quality of the generated recommendations comparable and often significantly better than those of other approaches employed for the comparison.

Image annotation is an important and challenging task when managing large image collections. In this paper, a fuzzy shape annotation approach for semi-automatic image annotation is presented. A fuzzy clustering process guided by partial supervision is applied to shapes represented by Fourier descriptors in order to derive a set of shape prototypes representative of a number of semantic categories. Next, prototypes are manually annotated by attaching textual labels related to semantic categories. Based on the labeled proto-types, a new shape is automatically labeled by associating a fuzzy set that provides membership degrees of the shape to all semantic categories. The proposed annotation approach provides an innovative indexing method for shape-based image retrieval. Indeed, shape prototypes represent an inter-mediate indexing level that allows a faster retrieval process since a query is matched against prototypes, instead of the whole shape database, resulting in a speed up of the retrieval. The proposed approach is tested on synthetic and real-word images in order to show its suitability.

In this work we propose the use of partially supervised fuzzy clustering to create a two-level indexing structure useful for enabling efficient shape retrieval. Similar shapes are grouped by a fuzzy clustering algorithm that embeds a partial supervision mechanism exploiting domain knowledge expressed in terms of a set of labeled shapes. After clustering, a set of prototypes representative of shape clusters is derived and used as indexing mechanism for retrieval. A shape query is matched against prototypes, instead of the whole shape database, and then shapes belonging to clusters for which prototype similarity is higher are returned. Experimental results obtained on two different datasets are presented to show the effectiveness of the proposed approach.

Automatic image annotation is an important and challenging task when managing large image collections. In this paper, we present an incremental approach for shape labeling, which is useful to image annotation when new sets of images are available during time. Every time new shape images are available, a semi-supervised fuzzy clustering algorithm is used to group shapes into a number of clusters by exploiting knowledge about classes expressed as a set of pre-labeled shapes. Each cluster is represented by a prototype that is manually labeled and used to annotate shapes. To capture the evolution of the image set, the previously discovered prototypes are added as pre-labeled shapes to the current shape set before clustering. The performance of the proposed incremental approach is evaluated on an image dataset from the fish domain, which is divided into chunks of data to simulate the progressive availability of shapes during time.

E-Government is becoming more attentive towards providing intelligent personalized services to citizens so that they can receive better services with less time and effort. This work presents an approach for inferring user segments that could be properly exploited to offer personalized services that better satisfy user needs and their expectations. User segments are derived starting from data that essentially describe demographic characteristics of users and that are gathered by questionnaires. A clustering process is performed on gathered data in order to derive user segments, i.e. groups of users sharing similar characteristics. Finally, for each derived segment, we define a user profile that summarizes characteristics shared by users belonging to the same segment. The suitability of the proposed approach is shown by providing results obtained on a case study.

The heterogeneous nature of the Web combined with the rapid diffusion of Web-based applications have made Web browsing an intricate activity for users. This has given rise to an urgent need for developing systems capable to assist and guide users during their navigational activity in the Web. Web Usage Mining (WUM) refers to the application of Data Mining techniques for the automatic discovery of meaningful usage patterns characterizing the browsing behavior of users, starting from access data collected from interactions of users with sites. The discovered patterns may be conveniently exploited in order to implement functionalities offering useful assistance to users. This chapter is mainly intended to provide an overview of the different stages involved in a general WUM process. As an example, a WUM approach is presented which is based on the use of fuzzy clustering to discovery user categories starting from usage patterns.