Probabilistic Relational Models Research Articles

Combining ontology and probabilistic models for the design of bio-based product transformation processes

This paper presents a workflow for the design of transformation processes using different kinds of expert’s knowledge. It introduces POND (Process and observation ONtology Discovery), a workflow dedicated to answer expert’s questions about processes. It addresses two main issues: (1) how to represent the processes inner complexity, and (2) how to reason about processes taking into account uncertainty and causality. First, we show how to use a semantic model, an ontology, and its associated data to answer some of the expert’s questions concerning the processes, using semantic web languages and technologies. Then, we describe how to learn a predictive model, to discover new knowledge and provide explicative models by integrating the semantic model into a probabilistic relational model. The result is a complete workflow able to extensively analyze transformation processes through all their granularity levels and answer expert’s questions about their domains. An example of this workflow is given on biocomposites manufacturing for food packaging.

On the Construction of Symmetries and Retaining Lifted Representations in Dynamic Probabilistic Relational Models

On the Construction of Symmetries and Retaining Lifted Representations in Dynamic Probabilistic Relational Models

A relational database model and algebra integrating fuzzy attributes and probabilistic tuples

Although there have been many fuzzy or probabilistic relational database models proposed for representing and handling imprecise and uncertain information of objects in real-world applications, models combining the relevance and strength of both fuzzy set theory and probability theory appear sporadic. In this paper, we propose a new fuzzy and probabilistic relational database model where the imprecision of an attribute value is represented by a fuzzy set and the uncertainty of a relational tuple is represented by a probability interval. The mass assignment theory is employed to deal with the challenge of integration and computation of both fuzzy sets and probabilities in the same model. The conjunction and disjunction strategies to combine imprecise and uncertain information are introduced. Then the fundamental concepts of the classical relational database model are extended and generalized in this new model. The syntax and semantics of the selection operation are formally defined. Finally, the other important algebraic operations on imprecise attributes and uncertain tuples are developed.

Concept Drift Detection in Dynamic Probabilistic Relational Models

Dynamic probabilistic relational models, which are factorized w.r.t. a full joint distribution, are used to cater for uncertainty and for relational and temporal aspects in real-world data. While these models assume the underlying temporal process to be stationary, real-world data often exhibits non-stationary behavior where the full joint distribution changes over time. We propose an approach to account for non-stationary processes w.r.t. to changing probability distributions over time, an effect known as concept drift. We use factorization and compact encoding of relations to efficiently detect drifts towards new probability distributions based on evidence.

RETRACTED ARTICLE: Sentiment topic sarcasm mixture model to distinguish sarcasm prevalent topics based on the sentiment bearing words in the tweets

Sentiment analysis as we all know is a developed field in which new features keeps on adding, but most of the time, on the internet people use sarcasm to convey their message which is very difficult to understand both by people and machines. Sarcastic statements are very complex as most of the time they sound in positive context if interpreted literally but actually the speaker mean the opposite of what they speak. Sarcasm detection is a subtask of opinion mining. The main intention behind sarcasm detection is to identify the user opinions or emotions expressed by the user in the written text. It plays a critical role in sentiment analysis by correctly identifying sarcastic or non sarcastic sentences. The sarcastic sentence has mixed polarity of both positive and negative words. Understanding sarcasm is quite a difficult and a challenging task even for humans as well as for machines. Various approaches for sarcasm detection are purely based on machine learning classifiers where training the classifier is based on simple lexical or dictionary based features. The objective of the work is to develop an unsupervised probabilistic relational model to identify sarcasm prevalent topics based on the sentiment distribution of the words in the tweets. The model estimates sentiment based topic level distribution. The model evaluation shows the sentiment associated words that do appear in the short text given the sentiment related label. The model outperforms the other baseline state of art Model for sarcasm detection as shown in the experimental result and it is very much suited for the prediction of sarcasm of a short tweet.

DARPA's Role in Machine Learning

Machine learning methods provide a way for artificial intelligence systems to learn from experience. This article describes four threads of machine learning research supported and guided by the Defense Advanced Research Projects Agency — probabilistic modeling for speech recognition, probabilistic relational models, the integration of multiple machine learning approaches into a task‐specific system, and neural network technology. These threads illustrate the Defense Advanced Research Projects Agency way of creating timely advances in a field.

Meta-Amortized Variational Inference and Learning

Despite the recent success in probabilistic modeling and their applications, generative models trained using traditional inference techniques struggle to adapt to new distributions, even when the target distribution may be closely related to the ones seen during training. In this work, we present a doubly-amortized variational inference procedure as a way to address this challenge. By sharing computation across not only a set of query inputs, but also a set of different, related probabilistic models, we learn transferable latent representations that generalize across several related distributions. In particular, given a set of distributions over images, we find the learned representations to transfer to different data transformations. We empirically demonstrate the effectiveness of our method by introducing the MetaVAE, and show that it significantly outperforms baselines on downstream image classification tasks on MNIST (10-50%) and NORB (10-35%).

Bayesian Networks for Data Integration in the Absence of Foreign Keys

In the era of open data, a single data source rarely contains all of the attributes we need for inference in specific applications. For example, a marketing department may aim to integrate retailer-specific purchase data with separate demographic data for purposes of targeted advertising – a capability not possible with either dataset alone. In this work, we address two key desiderata of an automated framework for probabilistic data integration over multiple data sources: (1) we require that each relational data source share at least one attribute with another relational data source, but we do not require these attributes to be foreign keys (e.g., attributes such as gender, age, and postal code are not foreign keys because they do not uniquely identify individuals in a data source) and (2) we require inference to be probabilistic to reflect inherent uncertainty in population-level predictions given the absence of foreign keys. While some frameworks such as Probabilistic Relational Models (PRMs) address point (2), they do not address point (1) since they rely on foreign keys to link tables. To achieve both desiderata simultaneously, we develop an automated framework to construct Bayesian networks for data integration capable of answering any probabilistic query spanning the attributes of multiple relational data sources. We demonstrate that our framework is able to closely approximate the inference of a global Bayesian network over a single relation that has been projected onto multiple local relations and further investigate properties of local relations such as the number of shared attributes and their cardinality to understand how these properties affect the quality of inference.

EXTENDING RELATIONAL DATABASE MODEL FOR UNCERTAIN INFORMATION

In this paper, we propose a new probabilistic relational database model, denote by PRDB, as an extension of the classical relational database model where the uncertainty of relational attribute values and tuples are respectively represented by finite sets and probability intervals. A probabilistic interpretation of binary relations on finite sets is proposed for the computation of their probability measures. The combination strategies on probability intervals are employed to combine attribute values and compute uncertain membership degrees of tuples in a relation. The fundamental concepts of the classical relational database model are extended and generalized for PRDB. Then, the probabilistic relational algebraic operations are formally defined accordingly in PRDB. In addition, a set of the properties of the algebraic operations in this new model also are formulated and proven.

Uncertainty Processing and Risk Monitoring in Construction Projects Using Hierarchical Probabilistic Relational Models

AbstractConstruction projects do not often reach their expected results regarding time, cost, and quality, due to the internal and external environment variations. Despite a substantial literature about risk management, no generic approach is proposed to represent construction project considering together technical and human dimensions or sustainability with their uncertainties. Modeling complex dynamical systems from heterogeneous pieces of knowledge varying in precision and reliability is a challenging task. This article proposes an innovative generic and versatile approach, based on the formalism of hierarchical probabilistic relational models to analyze and to propagate uncertainty in construction project regarding different levels of knowledge. The aim is to obtain a flexible, portable, and versatile model able to simulate the behavior of complex system's entities involved in any construction project at different levels of detail while taking uncertainty into account. To illustrate and highlight this approach, an academic example and a real case are proposed.

The complexity of Bayesian networks specified by propositional and relational languages

We examine the complexity of inference in Bayesian networks specified by logical languages. We consider representations that range from fragments of propositional logic to function-free first-order logic with equality; in doing so we cover a variety of plate models and of probabilistic relational models. We study the complexity of inferences when network, query and domain are the input (the inferential and the combined complexity), when the network is fixed and query and domain are the input (the query/data complexity), and when the network and query are fixed and the domain is the input (the domain complexity). We draw connections with probabilistic databases and liftability results, and obtain complexity classes that range from polynomial to exponential levels; we identify new languages with tractable inference, and we relate our results to languages based on plates and probabilistic relational models.

On the Coherence of Probabilistic Relational Formalisms.

There are several formalisms that enhance Bayesian networks by including relations amongst individuals as modeling primitives. For instance, Probabilistic Relational Models (PRMs) use diagrams and relational databases to represent repetitive Bayesian networks, while Relational Bayesian Networks (RBNs) employ first-order probability formulas with the same purpose. We examine the coherence checking problem for those formalisms; that is, the problem of guaranteeing that any grounding of a well-formed set of sentences does produce a valid Bayesian network. This is a novel version of de Finetti’s problem of coherence checking for probabilistic assessments. We show how to reduce the coherence checking problem in relational Bayesian networks to a validity problem in first-order logic augmented with a transitive closure operator and how to combine this logic-based approach with faster, but incomplete algorithms.

Analysing and modelling runtime architectural stability for self-adaptive software

Abstract With the increased dependence on software, there is a pressing need for engineering long-lived software. As architectures have a profound effect on the life-span of the software and the provisioned quality of service, stable architectures are significant assets. Architectural stability tends to reflect the success of the system in supporting continuous changes without phasing-out. For self-adaptive architectures, the behavioural aspect of stability is essential for seamless operation, to continuously keep the provision of quality requirements stable and prevent unnecessary adaptations that will risk degrading the system. In this paper, we introduce a systematic approach for analysing and modelling architectural stability. Specifically, we leverage architectural concerns and viewpoints to explicitly analyse stability attributes of the intended behaviour. Due to the probabilistic nature of systems’ behaviour, stability modelling is based on a probabilistic relational model for knowledge representation of stability multiple viewpoints. The model, empowered by the quantitative analysis of Bayesian networks, is capable to conduct runtime inference for reasoning about stability under runtime uncertainty. To illustrate the applicability and evaluate the proposed approach, we consider the case of cloud architectures. The results show that the approach increases the efficiency of the architecture in keeping the expected behaviour stable during runtime operation.

Open-Universe Weighted Model Counting

Weighted model counting (WMC) has recently emerged as an effective and general approach to probabilistic inference, offering a computational framework for encoding a variety of formalisms, such as factor graphs and Bayesian networks.The advent of large-scale probabilistic knowledge bases has generated further interest in relational probabilistic representations, obtained by according weights to first-order formulas, whose semantics is given in terms of the ground theory, and solved by WMC. A fundamental limitation is that the domain of quantification, by construction and design, is assumed to be finite, which is at odds with areas such as vision and language understanding, where the existence of objects must be inferred from raw data. Dropping the finite-domain assumption has been known to improve the expressiveness of a first-order language for open-universe purposes, but these languages, so far, have eluded WMC approaches. In this paper, we revisit relational probabilistic models over an infinite domain, and establish a number of results that permit effective algorithms. We demonstrate this language on a number of examples, including a parameterized version of Pearl's Burglary-Earthquake-Alarm Bayesian network.

Probabilistic relational models with class hierarchies to represent and simulate the multi-layer system of construction projects.

De nombreux facteurs et evenements tels que des erreurs humaines ou des aleas naturels peuvent conduire a l’echec des projets de construction (non-respect des delais, des couts…). En depit d'une documentation importante sur la gestion des risques, aucune approche holistique n'est proposee pour representer un projet de construction integrant aussi bien des dimensions techniques et humaines qu’envrionnementales. Dans cet article, un modele generique et holistique innovant, base sur le formalisme de modeles relationnels probabilistes avec hierarchies de classes (PRM-CHs), a ete developpe pour analyser et propager les risques dans les projets de construction. Ce modele est capable de (1) decrire et instancier tout type de projet de construction, (2) tenir compte de l'incertitude et (3) simuler et predire le comportement de chaque variable a differents niveaux de detail au cours du projet de construction. Afin d'illustrer cette approche, un cas reel simplifie (renovation de projet routier a Hue, Vietnam) est propose.

Probabilistic Inference on Multiple Normalized Genome-Wide Signal Profiles With Model Regularization.

Understanding genome-wide protein-DNA interaction signals forms the basis for further focused studies in gene regulation. In particular, the chromatin immunoprecipitation with massively parallel DNA sequencing technology (ChIP-Seq) can enable us to measure the in vivo genome-wide occupancy of the DNA-binding protein of interest in a single run. Multiple ChIP-Seq runs thus inherent the potential for us to decipher the combinatorial occupancies of multiple DNA-binding proteins. To handle the genome-wide signal profiles from those multiple runs, we propose to integrate regularized regression functions (i.e., LASSO, Elastic Net, and Ridge Regression) into the well-established SignalRanker and FullSignalRanker frameworks, resulting in six additional probabilistic models for inference on multiple normalized genome-wide signal profiles. The corresponding model training algorithms are devised with computational complexity analysis. Comprehensive benchmarking is conducted to demonstrate and compare the performance of nine related probabilistic models on the ENCODE ChIP-Seq datasets. The results indicate that the regularized SignalRanker models, in contrast to the original SignalRanker models, can demonstrate excellent inference performance comparable to the FullSignalRanker models with low model complexities and time complexities. Such a feature is especially valuable in the context of the rapidly growing genome-wide signal profile data in the recent years.

Behavioral determinants towards enhancing construction waste management: A Bayesian Network analysis

In this study, a Bayesian Network analysis is used to define causal behavioral determinants towards improving practices in construction waste management (CWM). For this purpose, a structured survey questionnaire was developed and administered to field workers at construction projects. The collected data was used to develop a probabilistic relational model with single- and multi-factor analysis to assess conditional probabilities underlying various determinants. The results indicate that behavior is highly influenced by attitude, past experience, and social pressure with 21, 20, and 10% higher chance of improvement by these factors, respectively. Behavior in CWM appears to be more sensitive to changes in personal factors such as attitude than corporate factors such as training. When simultaneously controlling all factors, the behavior is improved with personal factors by 9% more than with corporate factors. Additionally, it was found that the probability of having effective CWM practices on-site reaches 83% when workers have a positive attitude towards waste management, are well experienced in CWM practices, and are influenced by social pressure. Achieving this result also requires independency at work and availability of training sessions. The model raises the awareness of construction stakeholders about factors influencing workers’ behavior towards CWM and presents quantified strategies that increase the chances of minimizing the generation of construction waste. It can serve as a motivation and a decision support tool for adopting and implementing sustainable CWM practices.

Probabilistic relational model benchmark generation: Principle and application

The validation of any database mining methodology goes through an evaluation process where benchmarks availability is essential. In this paper, we aim to randomly generate relational database benchmarks that allow to check probabilistic dependencies among the attributes. We are particularly interested in Probabilistic relational models (PRMs). These latter extend Bayesian networks (BNs) to a relational data mining context that enable effective and robust reasoning about relational data structures. Even though a panoply of works have focused, separately, on Bayesian networks and relational databases random generation, no work has been identified for PRMs on that track. This paper provides an algorithmic approach allowing to generate random PRMs from scratch to cover the absence of generation process. The proposed method allows to generate PRMs as well as synthetic relational data from a randomly generated relational schema and a random set of probabilistic dependencies. This can be of interest for machine learning researchers to evaluate their proposals in a common framework, as for databases designers to evaluate the effectiveness of the components of a database management system.

Statistical Relational Artificial Intelligence: Logic, Probability, and Computation

An intelligent agent interacting with the real world will encounter individual people, courses, test results, drugs prescriptions, chairs, boxes, etc., and needs to reason about properties of these individuals and relations among them as well as cope with uncertainty. Uncertainty has been studied in probability theory and graphical models, and relations have been studied in logic, in particular in the predicate calculus and its extensions. This book examines the foundations of combining logic and probability into what are called relational probabilistic models. It introduces representations, inference, and learning techniques for probability, logic, and their combinations. The book focuses on two representations in detail: Markov logic networks, a relational extension of undirected graphical models and weighted first-order predicate calculus formula, and Problog, a probabilistic extension of logic programs that can also be viewed as a Turing-complete relational extension of Bayesian networks.

A probabilistic relational database model and algebra

This paper introduces a probabilistic relational database model, called PRDB, for representing and querying uncertain information of objects in practice. To develop the PRDB model, first, we represent the relational attribute value as a pair of probabilistic distributions on a set for modeling the possibility that the attribute can take one of the values of the set with a probability belonging to the interval which is inferred from the pair of probabilistic distributions. Next, on the basis representing such attribute values, we formally define the notions as the schema, relation, probabilistic functional dependency and probabilistic relational algebraic operations for PRDB. In addition, a set of the properties of the probabilistic relational algebraic operations in PRDB also are formulated and proven.