Reward models (RMs) are a crucial component in the alignment of large language models’ (LLMs) outputs with human values. RMs approximate human preferences over possible LLM responses to the same prompt by predicting and comparing reward scores. However, as they are typically modified versions of LLMs with scalar output heads, RMs are large black boxes whose predictions are not explainable. More transparent RMs would enable improved trust in the alignment of LLMs. In this work, we propose to use contrastive explanations to explain any binary response comparison made by an RM. Specifically, we generate a diverse set of new comparisons similar to the original one to characterise the RM’s local behaviour. The perturbed responses forming the new comparisons are generated to explicitly modify manually specified high-level evaluation attributes, on which analyses of RM behaviour are grounded. In quantitative experiments, we validate the effectiveness of our method for finding high-quality contrastive explanations. We then showcase the qualitative usefulness of our method for investigating global sensitivity of RMs to each evaluation attribute, and demonstrate how representative examples can be automatically extracted to explain and compare behaviours of different RMs. We see our method as a flexible framework for RM explanation, providing a basis for more interpretable and trustworthy LLM alignment.

Fine-tuning LLMs on tabular classification tasks can lead to the phenomenon of *fine-tuning multiplicity* where equally well-performing models make conflicting predictions on the same input. Fine-tuning multiplicity can arise due to variations in the training process, e.g., seed, weight initialization, minor changes to training data, etc., raising concerns about the reliability of Tabular LLMs in high-stakes applications such as finance, hiring, education, healthcare. Our work formalizes this unique challenge of fine-tuning multiplicity in Tabular LLMs and proposes a novel measure to quantify the consistency of individual predictions without expensive model retraining. Our measure quantifies a prediction's consistency by analyzing (sampling) the model's local behavior around that input in the embedding space. Interestingly, we show that sampling in the local neighborhood can be leveraged to provide probabilistic guarantees on prediction consistency under a broad class of fine-tuned models, i.e., inputs with sufficiently high local stability (as defined by our measure) also remain consistent across several fine-tuned models with high probability. We perform experiments on multiple real-world datasets to show that our local stability measure preemptively captures consistency under actual multiplicity across several fine-tuned models, outperforming competing measures.

Machine unlearning offers a practical alternative to avoid full model re-training by approximately removing the influence of specific user data. While existing methods certify unlearning via statistical indistinguishability from re-trained models, these guarantees do not naturally extend to model outputs when inputs are adversarially perturbed. In particular, slight perturbations of forget samples may still be correctly recognized by the unlearned model---even when a re-trained model fails to do so---revealing a novel privacy risk: information about the forget samples may persist in their local neighborhood. In this work, we formalize this vulnerability as residual knowledge and show that it is inevitable in high-dimensional settings. To mitigate this risk, we propose a fine-tuning strategy, named RURK, that penalizes the model’s ability to re-recognize perturbed forget samples. Experiments on vision benchmarks with deep neural networks demonstrate that residual knowledge is prevalent across existing unlearning methods and that our approach effectively prevents residual knowledge.

Although neural models have achieved remarkable performance, they still encounter doubts due to the intransparency. To this end, model prediction explanation is attracting more and more attentions. However, current methods rarely incorporate external knowledge and still suffer from three limitations: (1) Neglecting concept completeness. Merely selecting concepts may not sufficient for prediction. (2) Lacking concept fusion. Failure to merge semantically-equivalent concepts. (3) Difficult in manipulating model behavior. Lack of verification for explanation on original model. To address these issues, we propose a novel knowledge-aware neuron interpretation framework to explain model predictions for image scene classification. Specifically, for concept completeness, we present core concepts of a scene based on knowledge graph, ConceptNet, to gauge the completeness of concepts. Our method, incorporating complete concepts, effectively provides better prediction explanations compared to baselines. Furthermore, for concept fusion, we introduce a knowledge graph-based method known as Concept Filtering, which produces over 23% point gain on neuron behaviors for neuron interpretation. At last, we propose Model Manipulation, which aims to study whether the core concepts based on ConceptNet could be employed to manipulate model behavior. The results show that core concepts can effectively improve the performance of original model by over 26%.

The SHAP framework provides a principled method to explain the predictions of a model by computing feature importance. Motivated by applications in finance, we introduce the Top-k Identification Problem (TkIP) (and its ordered variant TkIP- O), where the objective is to identify the subset (or ordered subset for TkIP-O) of k features corresponding to the highest SHAP values with PAC guarantees. While any sampling-based method that estimates SHAP values (such as KernelSHAP and SamplingSHAP) can be trivially adapted to solve TkIP, doing so is highly sample inefficient. Instead, we leverage the connection between SHAP values and multi-armed bandits (MAB) to show that both TkIP and TkIP-O can be reduced to variants of problems in MAB literature. This reduction allows us to use insights from the MAB literature to develop sample-efficient variants of KernelSHAP and SamplingSHAP. We propose KernelSHAP@k and SamplingSHAP@k for solving TkIP; along with KernelSHAP-O and SamplingSHAP-O to solve the ordering problem in TkIP-O. We perform extensive experiments using several credit-related datasets to show that our methods offer significant improvements of up to 40× in sample efficiency and 39× in runtime.

Event relation extraction (ERE) is a critical and fundamental challenge for natural language processing. Existing work mainly focuses on directly modeling the entire document, which cannot effectively handle long-range dependencies and information redundancy. To address these issues, we propose a cluster-aware compression method for improving event relation extraction (TacoERE), which explores a compression-then-extraction paradigm. Specifically, we first introduce document clustering for modeling event dependencies. It splits the document into intra- and inter-clusters, where intra-clusters aim to enhance the relations within the same cluster, while inter-clusters attempt to model the related events at arbitrary distances. Secondly, we utilize cluster summarization to simplify and highlight important text content of clusters for mitigating information redundancy and event distance. We have conducted extensive experiments on both pre-trained language models, such as RoBERTa, and large language models, such as ChatGPT and GPT-4, on three ERE datasets, i.e., MAVEN-ERE, EventStoryLine and HiEve. Experimental results demonstrate that TacoERE is an effective method for ERE.

This paper proposes Progressive inference–a framework to explain the predictions of decoder-only transformer models trained to perform sequence classification tasks. Our work is based on the insight that the classification head of a decoder-only model can be used to make intermediate predictions by evaluating them at different points in the input sequence. Due to the masked attention mechanism used in decoder-only models, these intermediate predictions only depend on the tokens seen before the inference point, allowing us to obtain the model’s prediction on a masked input sub-sequence, with negligible computational overheads. We develop two methods to provide sub-sequence level attributions using this core insight. First, we propose Single Pass-Progressive Inference (SP-PI) to compute attributions by simply taking the difference between intermediate predictions. Second, we exploit a connection with Kernel SHAP to develop Multi Pass-Progressive Inference (MP-PI); this uses intermediate predictions from multiple masked versions of the input to compute higher-quality attributions that approximate SHAP values. We perform studies on several text classification datasets to demonstrate that our proposal provides better explanations compared to prior work, both in the single-pass and multi-pass settings.

We introduce a novel approach for detecting distribution shifts that negatively impact the performance of machine learning models in continuous production environments, which requires no access to ground truth data labels. It builds upon the work of Podkopaev and Ramdas [2022], who address scenarios where labels are available for tracking model errors over time. Our solution extends this framework to work in the absence of labels, by employing a proxy for the true error. This proxy is derived using the predictions of a trained error estimator. Experiments show that our method has high power and false alarm control under various distribution shifts, including covariate and label shifts and natural shifts over geography and time.

The Rashomon effect is a mixed blessing in responsible machine learning. It enhances the prospects of finding models that perform well in accuracy while adhering to ethical standards, such as fairness or interpretability. Conversely, it poses a risk to the credibility of machine decisions through predictive multiplicity. While recent studies have explored the Rashomon effect across various machine learning algorithms, its impact on gradient boosting---an algorithm widely applied to tabular datasets---remains unclear. This paper addresses this gap by systematically analyzing the Rashomon effect and predictive multiplicity in gradient boosting algorithms. We provide rigorous theoretical derivations to examine the Rashomon effect in the context of gradient boosting and offer an information-theoretic characterization of the Rashomon set. Additionally, we introduce a novel inference technique called RashomonGB to efficiently inspect the Rashomon effect in practice. On more than 20 datasets, our empirical results show that RashomonGB outperforms existing baselines in terms of improving the estimation of predictive multiplicity metrics and model selection with group fairness constraints. Lastly, we propose a framework to mitigate predictive multiplicity in gradient boosting and empirically demonstrate its effectiveness.

Data distillation and coresets have emerged as popular approaches to generate a smaller representative set of samples for downstream learning tasks to handle large-scale datasets. At the same time, machine learning is being increasingly applied to decision-making processes at a societal level, making it imperative for modelers to address inherent biases towards subgroups present in the data. While current approaches focus on creating fair synthetic representative samples by optimizing local properties relative to the original samples, their impact on downstream learning processes has yet to be explored. In this work, we present fair Wasserstein coresets ($\texttt{FWC}$), a novel coreset approach which generates fair synthetic representative samples along with sample-level weights to be used in downstream learning tasks. $\texttt{FWC}$ uses an efficient majority minimization algorithm to minimize the Wasserstein distance between the original dataset and the weighted synthetic samples while enforcing demographic parity. We show that an unconstrained version of $\texttt{FWC}$ is equivalent to Lloyd's algorithm for k-medians and k-means clustering. Experiments conducted on both synthetic and real datasets show that $\texttt{FWC}$: (i) achieves a competitive fairness-performance tradeoff in downstream models compared to existing approaches, (ii) improves downstream fairness when added to the existing training data and (iii) can be used to reduce biases in predictions from large language models (GPT-3.5 and GPT-4).

Interpreting the inner function of neural networks is crucial for the trustworthy development and deployment of these black-box models. Prior interpretability methods focus on correlation-based measures to attribute model decisions to individual examples. However, these measures are susceptible to noise and spurious correlations encoded in the model during the training phase (e.g., biased inputs, model overfitting, or misspecification). Moreover, this process has proven to result in noisy and unstable attributions that prevent any transparent understanding of the model’s behavior. In this paper, we develop a robust interventional-based method grounded by causal analysis to capture cause-effect mechanisms in pre-trained neural networks and their relation to the prediction. Our novel approach relies on path interventions to infer the causal mechanisms within hidden layers and isolate relevant and necessary information (to model prediction), avoiding noisy ones. The result is task-specific causal explanatory graphs that can audit model behavior and express the actual causes underlying its performance. We apply our method to vision models trained on classification tasks. On image classification tasks, we provide extensive quantitative experiments to show that our approach can capture more stable and faithful explanations than standard attribution-based methods. Furthermore, the underlying causal graphs express the neural interactions in the model, making it a valuable tool in other applications (e.g., model repair).

Feature selection is a crucial step in building machine learning models. This process is often achieved with accuracy as an objective, and can be cumbersome and computationally expensive for large-scale datasets. Several additional model performance characteristics such as fairness and robustness are of importance for model development. As regulations are driving the need for more trustworthy models, deployed models need to be corrected for model characteristics associated with responsible artificial intelligence. When feature selection is done with respect to one model performance characteristic (eg. accuracy), feature selection with secondary model performance characteristics (eg. fairness and robustness) as objectives would require going through the computationally expensive selection process from scratch. In this paper, we introduce the problem of feature reselection, so that features can be selected with respect to secondary model performance characteristics efficiently even after a feature selection process has been done with respect to a primary objective. To address this problem, we propose REFRESH, a method to reselect features so that additional constraints that are desirable towards model performance can be achieved without having to train several new models. REFRESH’s underlying algorithm is a novel technique using SHAP values and correlation analysis that can approximate for the predictions of a model without having to train these models. Empirical evaluations on three datasets, including a large-scale loan defaulting dataset show that REFRESH can help find alternate models with better model characteristics efficiently. We also discuss the need for reselection and REFRESH based on regulation desiderata.

Similarity functions measure how comparable pairs of elements are, and play a key role in a wide variety of applications, e.g., notions of Individual Fairness abiding by the seminal paradigm of Dwork et al., as well as Clustering problems. However, access to an accurate similarity function should not always be considered guaranteed, and this point was even raised by Dwork et al. For instance, it is reasonable to assume that when the elements to be compared are produced by different distributions, or in other words belong to different ``demographic'' groups, knowledge of their true similarity might be very difficult to obtain. In this work, we present an efficient sampling framework that learns these across-groups similarity functions, using only a limited amount of experts' feedback. We show analytical results with rigorous theoretical bounds, and empirically validate our algorithms via a large suite of experiments.

Conventional convolution neural networks (CNNs) trained on narrow Field-of-View (FoV) images are the state-of-the art approaches for object recognition tasks. Some methods proposed the adaptation of CNNs to ultra-wide FoV images by learning deformable kernels. However, they are limited by the Euclidean geometry and their accuracy degrades under strong distortions caused by fisheye projections. In this work, we demonstrate that learning the shape of convolution kernels in non-Euclidean spaces is better than existing deformable kernel methods. In particular, we propose a new approach that learns deformable kernel parameters (positions) in hyperbolic space. FisheyeHDK is a hybrid CNN architecture combining hyperbolic and Euclidean convolution layers for positions and features learning. First, we provide intuition of hyperbolic space for wide FoV images. Using synthetic distortion profiles, we demonstrate the effectiveness of our approach. We select two datasets - Cityscapes and BDD100K 2020 - of perspective images which we transform to fisheye equivalents at different scaling factors (analogue to focal lengths). Finally, we provide an experiment on data collected by a real fisheye camera. Validations and experiments show that our approach improves existing deformable kernel methods for CNN adaptation on fisheye images.

Relational Graph Convolutional Networks (RGCNs) are commonly used on Knowledge Graphs (KGs) to perform black box link prediction. Several algorithms have been proposed to explain their predictions. Evaluating performance of explanation methods for link prediction is difficult without ground truth explanations. Furthermore, there can be multiple explanations for a given prediction in a KG. No dataset exists where observations have multiple ground truth explanations to compare against. Additionally, no standard scoring metrics exist to compare predicted explanations against multiple ground truth explanations. We propose and evaluate a method, including a dataset, to benchmark explanation methods on the task of explainable link prediction using RGCNs.

Relational Graph Convolutional Networks (RGCNs) are commonly used on Knowledge Graphs (KGs) to perform black box link prediction. Several algorithms, or explanation methods, have been proposed to explain their predictions. Evaluating performance of explanation methods for link prediction is difficult without ground truth explanations. Furthermore, there can be multiple explanations for a given prediction in a KG. No dataset exists where observations have multiple ground truth explanations to compare against. Additionally, no standard scoring metrics exist to compare predicted explanations against multiple ground truth explanations. In this paper, we introduce a method, including a dataset (FrenchRoyalty-200k), to benchmark explanation methods on the task of link prediction on KGs, when there are multiple explanations to consider. We conduct a user experiment, where users score each possible ground truth explanation based on their understanding of the explanation. We propose the use of several scoring metrics, using relevance weights derived from user scores for each predicted explanation. Lastly, we benchmark this dataset on state-of-the-art explanation methods for link prediction using the proposed scoring metrics.

Visual Question Answering (VQA) remains algorithmically challenging while it is effortless for humans.Humans combine visual observations with general and commonsense knowledge to answer a question about a given image.In this paper, we address the problem of incorporating general knowledge into VQA models while leveraging the visual information.We propose a model that captures the interactions between objects in a visual scene and entities in an external knowledge source.Our model is a graph-based approach that combines scene graphs with concept graphs, which learns a question-adaptive graph representation of related knowledge instances.We use Graph Attention Networks to set higher importance to key knowledge instances that are mostly relevant to each question.We exploit ConceptNet as the source of general knowledge and evaluate the performance of our model on the challenging OK-VQA dataset.

Zero-shot learning (ZSL) is a popular research problem that aims at predicting for those classes that have never appeared in the training stage by utilizing the inter-class relationship with some side information. In this study, we propose to model the compositional and expressive semantics of class labels by an OWL (Web Ontology Language) ontology, and further develop a new ZSL framework with ontology embedding. The effectiveness has been verified by some primary experiments on animal image classification and visual question answering.

Feature selection is a crucial step in the conception of Machine Learning models, which is often performed via datadriven approaches that overlook the possibility of tapping into the human decision-making of the model’s designers and users. We present a human-in-the-loop framework that interacts with domain experts by collecting their feedback regarding the variables (of few samples) they evaluate as the most relevant for the task at hand. Such information can be modeled via Reinforcement Learning to derive a per-example feature selection method that tries to minimize the model’s loss function by focusing on the most pertinent variables from a human perspective. We report results on a proof-of-concept image classification dataset and on a real-world risk classification task in which the model successfully incorporated feedback from experts to improve its accuracy.

Flight delays impact airlines, airports and passengers. Delay prediction is crucial during the decision-making process for all players in commercial aviation, and in particular for airlines to meet their on-time performance objectives. Although many machine learning approaches have been experimented with, they fail in (i) predicting delays in minutes with low errors (less than 15 minutes), (ii) being applied to small carriers i.e., low cost companies characterized by a small amount of data. This work presents a Long Short-Term Memory (LSTM) approach to predicting flight delay, modeled as a sequence of flights across multiple airports for a particular aircraft throughout the day. We then suggest a transfer learning approach between heterogeneous feature spaces to train a prediction model for a given smaller airline using the data from another larger airline. Our approach is demonstrated to be robust and accurate for low cost airlines in Europe.

Transfer learning aims at building robust prediction models by transferring knowledge gained from one problem to another. In the semantic Web, learning tasks are enhanced with semantic representations. We exploit their semantics to augment transfer learning by dealing with when to transfer with semantic measurements and what to transfer with semantic embeddings. We further present a general framework that integrates the above measurements and embeddings with existing transfer learning algorithms for higher performance. It has demonstrated to be robust in two real-world applications: bus delay forecasting and air quality forecasting.

This paper proposes a congestion balancing parking guidance system that suggests to a driver a sequence of streets to follow around the desired destination with the aim to reduce the total distance that is travelled while searching for a free parking spot. The system requires only limited infrastructure information, and neither requires parking spaces to be instrumented, nor vehicles to communicate with each other. Specifically, the system utilizes parking vacancy information on each street. The system also accounts for the added cost of not finding a free space, which is typically expressed as the additional distance that needs to be travelled to find an available parking spot. To avoid local congestion, different drivers respond to different suggestions based on a probability distribution that considers the total distance that needs to be travelled. A mobility simulator is used to model the searching behaviors of vehicles for parking spaces with and without the smart parking algorithm and experimental results are provided using the road network of the city of Dublin, Ireland.

Data stream learning has been largely studied for extracting knowledge structures from continuous and rapid data records. In the semantic Web, data is interpreted in ontologies and its ordered sequence is represented as an ontology stream. Our work exploits the semantics of such streams to tackle the problem of concept drift i.e., unexpected changes in data distribution, causing most of models to be less accurate as time passes. To this end we revisited (i) semantic inference in the context of supervised stream learning, and (ii) models with semantic embeddings. The experiments show accurate prediction with data from Dublin and Beijing.

Explanatory diagnosis of an ontology stream aims to explain the changes hidden in the ontology stream by a sequence of actions. In this paper, we present a framework for explanatory diagnosis of an ontology stream, which allows the actions to be uncertain. In order to capture the semantics of actions, we introduce a new update operator and effect-guided bold-repair. By combining these operators with a query mechanism of description logicssupporting inconsistency-tolerant semantics, we present a formal definition for the explanatory diagnosis problem of ontology streams.

Most service composition approaches rely on top-down decomposition of a problem and AI-style planning to assemble service components into a meaningful whole, impeding reuse and flexibility. In this article, we propose an approach that starts from declarative knowledge about the semantics of individual service components and algorithmically constructs a full-blown service orchestration process that supports sequence, choice, and parallelism. The output of our algorithm can be mapped directly into a number of service orchestration languages such as OWL-S and BPEL. The approach consists of two steps. First, semantic links specifying data dependencies among the services are derived and organized in a flexible network. Second, based on a user request indicating the desired outcomes from the composition, an executable composition is constructed from the network that satisfies the dependencies. The approach is unique in producing complex compositions out of semantic links between services in a flexible way. It also allows reusing knowledge about semantic dependencies in the network to generate new compositions through new requests and modification of services at runtime. The approach has been implemented in a prototype that outperforms related composition prototypes in experiments.

Deductive reasoning and inductive learning are the most common approaches for deriving knowledge. In real world applications when data is dynamic and incomplete, especially those exposed by sensors, reasoning is limited by dynamics of data while learning is biased by data incompleteness. Therefore discovering consistent knowledge from incomplete and dynamic data is a challenging open problem. In our approach the semantics of data is captured through ontologies to empower learning (mining) with (Description Logics) reasoning. Consistent knowledge discovery is achieved by applying generic, significative, representative association semantic rules. The experiments have shown scalable, accurate and consistent knowledge discovery with data from Dublin.

Many various types of sensors coming from different complex devices collect data from a city. Their underlying data representation follows specific manufacturer specifications that have possibly incomplete descriptions (in ontology) alignments. This paper addresses the problem of determining accurate and complete matching of ontologies given some common descriptions and their pre-determined high level alignments. In this context the problem of ontology matching consists of automatically determining all matching given the latter alignments, and manually verifying the matching results. Especially for applications where it is crucial that ontologies are matched correctly the latter can turn into a very time-consuming task for the user. This paper tackles this challenge and addresses the problem of computing the minimum number of user inputs needed to verify all matchings. We show how to represent this problem as a reasoning problem over a bipartite graph and how to encode it over pseudo Boolean constraints. Experiments show that our approach can be successfully applied to real-world data sets.

Diagnosis, or the process of identifying the nature and cause of an anomaly in an ontology, has been largely studied by the Semantic Web community. In the context of ontology stream, diagnosis results are not captured by a unique fixed ontology but numerous time-evolving ontologies. Thus any anomaly can be diagnosed by a large number of different explana- tions depending on the version and evolution of the ontology. We address the problems of identifying, representing, exploiting and exploring the evolution of diagnoses representations. Our approach consists in a graph-based representation, which aims at (i) efficiently organizing and linking time-evolving di- agnoses and (ii) being used for scalable exploration. The ex- periments have shown scalable diagnoses exploration in the context of real and live data from Dublin City.

Data captured in OWL ontologies is generally considered to be more prone to changes than the schema in many situations. Such changes often necessitate consistency checking over the resulting ontologies in order to maintain coherent knowledge, specifically in dynamic settings. In this paper, we present an approach to check the consistency over an evolving ontology resulting from data insertions and deletions, given by some expressive underlying Description Logic dialect. The approach, assuming an initially consistent ontology, works by syntactically identifying "relevant" and representative parts of the data for the given updates, i.e., the part that may contribute to subsequent consistency checking. Our approach has demonstrated its efficacy in checking consistency over large and real-world ontologies and outperforms existing approaches in several circumstances.

Sensor systems are constantly growing in all application areas and become elements of our environment. Semantic Sensor Networks (SSN) support this development and provide standardized semantic access for reasoning on this information. Unfortunately they do not model internal system knowledge or simple correlations between sensors and hence they cannot be used to automatically perform analytics tasks based on sensor data only. We show how SSN ontology can be extended and demonstrate its benefits for the task of diagnosing smart building problems using real-world data.

Diagnosis is the task of explaining abnormal behaviors of systems like telecommunication, transportation or energy systems. Given a sequence of observations the problem is to determine, online, all faults that are in line with these observations. Many approaches tackle this problem but they either require domain expertise or a formal description of how observations and faults are connected. This limits their scope to the diagnosis of well-understood faults. We address the problem of diagnosing faults that may occur for the first time and present a new diagnosis approach that integrates techniques for analyzing semantic descriptions of observations and faults.

This paper presents STAR-CITY, a system supporting semantic traffic analytics and reasoning for city. STAR-CITY, which integrates (human and machine-based) sensor data using variety of formats, velocities and volumes, has been designed to provide insight on historical and real-time traffic conditions, all supporting efficient urban planning. Our system demonstrates how the severity of road traffic congestion can be smoothly analyzed, diagnosed, explored and predicted using semantic web technologies. We present how semantic diagnosis and predictive reasoning, both using and interpreting semantics of data to deliver useful, accurate and consistent inferences, have been exploited and adapted systematized in an intelligent user interface. Our prototype of semantics-aware traffic analytics and reasoning, experimented in Dublin City Ireland, works and scales efficiently with historical together with real live and heterogeneous stream data.

The advent of real-time traffic streaming offers users the opportunity to visualise current traffic conditions and congestion information. However, real-time information highlighting the underlying reason for tail-backs remains largely unexplored. Broken traffic lights, an accident, a large concert, or road-works reveal important information for citizens and traffic operators alike. Providing such information in real-time requires intelligent mechanisms and user interfaces in order to (i) harness heterogeneous data sources (volume, velocity, variety, veracity) and (ii) make derived knowledge consumable so users can visualize traffic conditions and congestion information making better routing decisions while travelling. This work focuses on surfacing relevant information and explaining the underlying reasons behind traffic conditions. To this end, static data from event providers, planned road works together with dynamically emerging events such as a traffic accidents, localized weather conditions or unplanned obstructions are captured through social media to provide users real-time feedback to highlight the causes of traffic congestion.

The current proliferation of software services means users should be supported when selecting one service out of the many which meet their needs. Recommender Systems provide such support for selecting products and conventional services, yet their direct application to software services is not straightforward, because of the current scarcity of available user feedback, and the need to fine-tune software services to the context of intended use. In this article, we address these issues by proposing a semantic content-based recommendation approach that analyzes the context of intended service use to provide effective recommendations in conditions of scarce user feedback. The article ends with two experiments based on a realistic set of semantic services. The first experiment demonstrates how the proposed semantic content-based approach can produce effective recommendations using semantic reasoning over service specifications by comparing it with three other approaches. The second experiment demonstrates the effectiveness of the proposed context analysis mechanism by comparing the performance of both context-aware and plain versions of our semantic content-based approach, benchmarked against user-performed selection informed by context.

Recently, ontology stream reasoning has been introduced as a multidisciplinary approach, merging synergies from Artificial Intelligence, Database and World-Wide-Web to reason on semantics-augmented data streams, thus a way to answering questions on real time events. However existing approaches do not consider stream change diagnosis i.e., identification of the nature and cause of changes, where explaining the logical connection of knowledge and inferring insight on time changing events are the main challenges. We exploit the Description Logics (DL)-based semantics of streams to tackle these challenges. Based on an analysis of stream behavior through change and inconsistency over DL axioms, we tackled change diagnosis by determining and constructing a comprehensive view on potential causes of inconsistencies. We report a large-scale evaluation of our approach in the context of live stream data from Dublin City Council.

Most service composition approaches rely on top down decomposition of a problem and AI-style planning to assemble services into a meaningful whole, impeding reuse and flexibility. In contrast, our approach starts from declarative knowledge about the semantics of individual services and constructs a full-blown orchestration process that supports sequence, choice and parallelism. The approach, which is able to generate OWL-S and WS-BPEL based description, is unique in producing complex compositions out of semantic links between services in a flexible way. It also allows reusing knowledge about semantic dependencies in the network to generate new compositions through new requests and modification of services at run-time.

Nowadays web users have clearly expressed their wishes to receive and interact with personalized services directly. However, existing approaches, largely syntactic content-based, fail to provide robust, accurate and useful personalized services to its users. Towards such an issue, the semantic web provides technologies to annotate and match services’ descriptions with users’ features, interests and preferences, thus allowing for more efficient access to services and more generally information. The aim of our work, part of service personalization, is on automated instantiation of services which is crucial for advanced usability i.e., how to prepare and present services ready to be executed while limiting useless interactions with users? We introduce the constructive Description Logics reasoning join and couple it with concept abduction to i) identify useful parts of users profiles that satisfy services requirements and ii) compute the description required by a service to be executed but not provided by users profiles.

Nowadays web users have clearly expressed their wishes to receive and interact with personalized services directly. However, existing approaches, largely syntactic content-based, fail to provide robust, accurate and useful personalized services to its users. Towards such an issue, the semantic web provides technologies to annotate and match services' descriptions with users' features, interests and preferences, thus allowing for more efficient access to services and more generally information. The aim of our work, part of service personalization, is on automated instantiation of services which is crucial for advanced usability i.e., how to prepare and present services ready to be executed while limiting useless interactions with users? To this end, we exploit Description Logics reasoning through semantic matching to (i) identify useful parts of a user profile that satisfy services requirements (i.e., input parameters) and (ii) compute the description required by a service to be executed but not provided by the user profile. Our approach, part of the EC-funded project SOA4All, was evaluated on its applicability in real world scenarios with end-users.

Automation of web service composition is one of the most interesting challenges facing the semantic web today. Despite approaches which are able to infer partial order on services, data flow (i.e., the way data is exchanged among services) remains implicit and difficult to be inferred and automatically generated. Since web services have been enhanced with formal semantic descriptions, it becomes conceivable to exploit and reason on their semantic links (i.e., semantic matching between their functional output and input parameters) to infer data flow. Our approach has been directed to meet the main challenges facing the latter problem i.e., how to effectively i) guarantee whether a data flow is well-formed and ii) infer data flow between services based on their Description Logics (DL) descriptions. To this end, we apply constructive DL reasoning abduction, contraction and introduce the non standard DL reasoning join to model and infer data flow in compositions. The preliminary evaluation results showed high efficiency and effectiveness of the proposed approach.

The current proliferation of software services means users should be supported when selecting one service out of the many which meet consumer's needs. Recommender Systems provide such support for selecting products, yet their direct application to software services is not straightforward. In this paper, we derive three requirements for software service recommender systems and then propose a hybrid recommendation approach to address these requirements and provide effective recommendations in conditions of scarce user feedback. The approach combines semantic Content-based reasoning and context-dependent Collaborative Filtering. The paper ends with the experiments based on a realistic set of semantic services against existing approaches, demonstrating how our approach can produce effective recommendation using semantic reasoning over service specifications.

Nowadays web users have clearly expressed their wishes to receive and interact with personalized services directly. However, existing approaches, largely syntactic content-based, fail to provide robust, accurate and useful personalized services to its users. Towards such an issue, the semantic web provides technologies to annotate and match services' descriptions with users' features, interests and preferences, thus allowing for more efficient access to services and more generally information. The aim of our work, part of service personalization, is on automated instantiation of services which is crucial for advanced usability i.e., how to prepare and present services ready to be executed while limiting useless interactions with users? To this end, we exploit Description Logics reasoning through semantic matching to (i) identify useful parts of a user profile that satisfy services requirements (i.e., input parameters) and (ii) compute the description required by a service to be executed but not provided by the user profile. Our approach, part of the EC-funded project SOA4All, was evaluated on its applicability in real world scenarios with end-users.

Ranking and optimization of web service compositions represent challenging areas of research with significant implications for the realization of the “Web of Services” vision. “Semantic web services” use formal semantic descriptions of web service functionality and interface to enable automated reasoning over web service compositions. To judge the quality of the overall composition, for example, we can start by calculating the semantic similarities between outputs and inputs of connected constituent services, and aggregate these values into a measure of semantic quality for the composition. This paper takes a specific interest in combining semantic and nonfunctional criteria such as quality of service (QoS) to evaluate quality in web services composition. It proposes a novel and extensible model balancing the new dimension of semantic quality (as a functional quality metric) with a QoS metric, and using them together as ranking and optimization criteria. It also demonstrates the utility of Genetic Algorithms to allow optimization within the context of a large number of services foreseen by the “Web of Services” vision. We test the performance of the overall approach using a set of simulation experiments, and discuss its advantages and weaknesses.

Web service composition enhanced by semantic technologies is currently one of the most hyped and addressed issues in Service Oriented Computing. This work focuses on both i) conditional composition, i.e., how to automate the composition of services t

Automated web service composition has been tackled from different directions and to different purposes. In addition, most of the approaches address the composition problem with under specified requirements, returning compositions models that do not necessarily satisfy and fulfill end-users objectives. Satisfying the latter objectives is a difficult problem, especially from scratch, which requires stronger requirements and a further step of integration with service-based components in order to make service oriented computing and service composition a reality. In this work, we address this issue by presenting an innovative and integrated approach to service composition which consists of i) an automatic template process generator, that is able to generate abstract process templates and their hierarchy from past executions; ii) a novel and scalable approach to AI parametric-design techniques using a multi agent approach to configure and adapt services processes, heavily relying on the latter set of abstract process templates; iii) an optimization process that maximizes the overall quality of final compositions. Finally, we compare the scalability of these components with some experiments.

Optimizating semantic Web service compositions is known to be NP-hard, so most approaches restrict the number of services and offer poor scalability. We address the scalability issue by selecting compositions which satisfy a set of constraints rather than attempting to produce an optimal composition. Firstly, we define constraints within an innovative and extensible quality model designed to balance semantic fit (or functional quality) with quality of service (QoS) metrics. The semantic fit criterion evaluates the quality of semantic links between the semantic description of Web services parameters, whilst QoS focuses on non-functional criteria of services. Coupling these criteria allows us to further constrain and select valid compositions. To allow the use of this model in the context of millions of services as foreseen by the strategic EC-funded project SOA4All, we i) formulate the selection problem as a constraint satisfaction problem and ii) test the use of a stochastic search method. Finally we compare the latter with state-of-the-art approaches.

Automated composition of Web services or the process of forming new value-added Web services is one of the most promising challenges facing the Semantic Web today. Semantics enables Web service to describe capabilities together with their processes, hence one of the key elements for the automated composition of Web services. In this paper, we focus on the functional level of Web services i.e. services are described according to some input, output parameters semantically enhanced by concepts in a domain ontology. Web service composition is then viewed as a composition of semantic links wherein the latter links refer to semantic matchmaking between Web service parameters (i.e. outputs and inputs) in order to model their connection and interaction. The key idea is that the matchmaking enables, at run time, finding semantic compatibilities among independently defined Web service descriptions. By considering such a level of composition, a formal model to perform the automated composition of Web services i.e. Semantic Link Matrix, is introduced. The latter model is required as a starting point to apply problem-solving techniques such as regression (or progression)-based search for Web service composition. The model supports a semantic context in order to find correct, complete, consistent and robust plans as solutions. In this paper, an innovative and formal model for an AI (Artificial Intelligence) planning-oriented composition is presented. Our system is implemented and interacting with Web services which are dedicated to Telecom scenarios. The preliminary evaluation results showed high efficiency and effectiveness of the proposed approach.

Automation of Web service composition is one of the most interesting challenges facing the Semantic Web today. Since Web services have been enhanced with formal semantic descriptions, it becomes conceivable to exploit causal links i.e., semantic matching between their functional parameters (i.e., outputs and inputs). The semantic quality of causal links involved in a composition can be then used as a innovative and distinguishing criterion to estimate its overall semantic quality. Therefore non functional criteria such as quality of service (QoS) are no longer considered as the only criteria to rank compositions satisfying the same goal. In this paper we focus on semantic quality of causal link based semantic Web service composition. First of all, we present a general and extensible model to evaluate quality of both elementary and composition of causal links. From this, we introduce a global causal link selection based approach to retrieve the optimal composition. This problem is formulated as an optimization problem which is solved using efficient integer linear programming methods. The preliminary evaluation results showed that our global selection based approach is not only more suitable than the local approach but also outperforms the naive approach.

Automation of Web service composition is one of the most interesting challenges facing the Service Oriented Computing today. From this challenge, many issues such as control flow, data flow, verification, execution monitoring, or recovery actions (e.g., compensation) follows. In this paper we focus on automated data flow in Web service composition. The semantic Web, as an evolving extension of the current Web, seems a key initiative to overcome the latter issue. However, even if some approaches focus on discovering potential semantic connections between Web services, few or none of these tackle implementations issues related to XML messages management at syntactic level. In this direction we present an approach for performing automated data flow in Web service composition by i) exploiting semantic matchmaking between Web service parameters (i.e., outputs and inputs) to enable their connection and interaction, and ii) adapting XML database solutions, specifically XML Schema mapping, to perform syntactic data transformation and integration of exchanged messages. Our system is implemented and interacting with Web services dedicated on a Telecom scenario. The preliminary evaluation results showed not only high efficiency and effectiveness of the proposed approach but also complementarity of the semantic matchmaking and syntactic mapping to achieving data flow in Web service composition.

We claim that a key feature for correct and effective web service composition, and one that has largely been ignored,is the joint consideration of (semantic) causal links and causal laws, respectively in area of Description Logics (DL) and AI planning. In this paper we propose a means of specifying both causal links and laws into web service composition by integrating DL reasoning and Situation Calculus. To this end an augmented and adapted version of the logic programming language Golog i.e., sclGolog is presented as a natural formalism not only for reasoning about the latter links and laws, but also for automatically composing services. sclGolog operates as an offline interpreter that supports n-ary sensing actions to retrieve conditional compositions of services. Lastly sclGolog has been implemented and tested in the context of Telecommunication scenarios.

Web services are conveniently advertised and published based on (stateless) functional descriptions, while they are usually realized as (stateful) processes. Therefore, the automated enactment of complex Web services on the basis of pre-existing ones requires the ability to handle services described at very different abstraction levels. This is the main reason behind the current lack of approaches capable to perform automated end-to-end composition, starting from semantic requirements to obtain executable orchestrations of stateful processes. In this paper we achieve such a challenging goal, by modularly integrating a range of incrementally more complex techniques that cover the necessary discovery and composition phases. By gradually bridging the gap between the high-level requirements and the concrete realization of services, our architecture manages sensibly the complexity of the problem: incrementally more complex techniques are provided with incrementally more focused input. The tests of our architecture on a deployed scenario witness the functionality of the platform and its integrability with standard service engines.

The semantic web promises to bring automation to the areas of web service selection, discovery, composition, invocation. In this paper we introduce a means of facilitating automation of web service composition by exploiting semantic matchmaking between web service parameters (i.e., outputs and inputs) to enable their connections and interactions. The idea is that matchmaking functions are key components to find semantic compatibilities among independently web service descriptions. To this end, our approach extends existing methods (exact, plug-in, subsume, intersection and fail) with concept abduction to provide explanations of misconnections between web services. From this we generate web service compositions that realize the goal, discovering and satisfying semantic connections between Web services. Moreover a process of relaxing the hard constraints is introduced in case the composition process failed. Our system is implemented and interacting with web services dedicated on a France Telecom scenario.