ACM Transactions on Intelligent Systems and Technology (TIST), 2014
Recommender systems are quickly becoming ubiquitous in applications such as e-commerce, social me... more Recommender systems are quickly becoming ubiquitous in applications such as e-commerce, social media channels, and content providers, among others, acting as an enabling mechanism designed to overcome the information overload problem by improving browsing and consumption experience. A typical task in many recommender systems is to output a ranked list of items, so that items placed higher in the rank are more likely to be interesting to the users. Interestingness measures include how accurate, novel, and diverse are the suggested items, and the objective is usually to produce ranked lists optimizing one of these measures. Suggesting items that are simultaneously accurate, novel, and diverse is much more challenging, since this may lead to a conflicting-objective problem, in which the attempt to improve a measure further may result in worsening other measures. In this article, we propose new approaches for multiobjective recommender systems based on the concept of Pareto efficiency—a...
Na construcao de uma aplicacao de aprendizado de maquina, um especialista define seu objetivo, de... more Na construcao de uma aplicacao de aprendizado de maquina, um especialista define seu objetivo, determina, baseado em algumas hipoteses, os dados que possuem uma relacao causal com o objetivo, seleciona o modelo que melhor se adequa a suas hipoteses e dados, realiza experimentos e analisa a qualidade da solucao. Na fase de analise, uma propriedade fundamental do modelo e a interpretabilidade. Em alguns dominios de aplicacao, como medica ou de negocios, a interpretabilidade e tida como diferencial. Para que um modelo seja interpretavel e indicado que ele possua poucos atributos e siga o principio da parcimonia no qual, dentre explicacoes equivalentes, as mais simples sao preferiveis. Esse principio tem se mostrado adequado para classificadores baseados em regras de associacao, nos quais a quantidade de regras utilizadas podem ser substancialmente reduzidas utilizando-se representacoes condensadas, como os conjuntos maximos ou fechados. Porem, a quantidade restante de regras ainda e gr...
The Occam's Razor principle has become the basis for many Machine Learning algorithms, under ... more The Occam's Razor principle has become the basis for many Machine Learning algorithms, under the interpretation that the classifier should not be more complex than necessary. Recently, this principle has shown to be well suited to associative classifiers, where the number of rules composing the classifier can be substantially reduced by using condensed representations such as maximal or closed rules. While it is shown that such a decrease in the complexity of the classifier (usually) does not compromise its accuracy, the number of remaining rules is still larger than necessary and making it hard for experts to interpret the corresponding classifier. In this paper we propose a much more aggressive filtering strategy, which decreases the number of rules within the classifier dramatically without hurting its accuracy. Our strategy consists in evaluating each rule under different statistical criteria, and filtering only those rules that show a positive balance between all the criter...
Performance modeling is a central task in the capacity management of server platforms. Traditiona... more Performance modeling is a central task in the capacity management of server platforms. Traditional performance models are created for transactional workloads (purely open models), batch or interactive (purely closed models). However, many real Web applications experience session-based workloads that exhibit a partially-open behavior, which includes components from either open or closed model, particularly with respect to response time. In this paper, we developed an analytical model for performance that captures the main aspects of partially-open workloads of multi-layered Web platforms and compared it with the open and closed models by simulation. Resumo. A modelagem de desempenhoé uma tarefa central para o gerenciamento de capacidade de plataformas servidoras. Modelos de desempenho tradicionais são derivados para cargas de trabalho transacionais (modelos puramente abertos), batch ou interativas (modelos puramente fechados). Entretanto, várias aplicações reais da Web consistem em cargas baseadas em sessões com comportamento semi-aberto que incluem componentes dos modelos abertos e fechados, particularmente em relação ao tempo de resposta. Neste artigo, desenvolvemos um modelo analítico de desempenho que captura aspectos de cargas semi-abertas para plataformas Web multi-camadas e analisamos seus compromissos em relação aos modelos fechado e aberto por meio de simulação.
In this work we present an extension of our previous capacity management framework making it robu... more In this work we present an extension of our previous capacity management framework making it robust to security attacks. In a scenario where multiple applications are hosted in a shared infrastructure, our performance and optimization models are extended to minimize the impact of security attacks on the provider's revenue and applications' QoS. A number of scenarios and strategies based on dynamic SLA contracts are designed to help uncover the main tradeoffs considering both the provider's interests and the customer's interests.
Capacity management of a hosting infrastructure has traditionally focused only on performance goa... more Capacity management of a hosting infrastructure has traditionally focused only on performance goals. However, the quality of service provided to the hosted applications and the revenues achieved by the provider, depend also on other aspects, such as security and energy constraints. This paper extends our self-adaptive SLA-driven capacity management solution to capture key performance and cost tradeoffs that arise when operating under security attacks and energy constraints. A number of scenarios and strategies based on dynamic SLA contracts are designed to help uncover the main tradeoffs considering both the provider's interests and the customers's interests.
Capacity management of a hosting infrastructure has traditionally focused only on performance goa... more Capacity management of a hosting infrastructure has traditionally focused only on performance goals. However, the quality of service provided to the hosted applications, and ultimately the revenues achieved by the provider, depend also on other aspects, such as security and energy constraints. This paper extends our self-adaptive SLA-driven capacity management solution to capture, in an unified framework, key performance and cost tradeoffs that arise when operating under security attacks and energy constraints. A number of scenarios and strategies based on dynamic SLA contracts are designed to help uncover, via simulation experiments, the main tradeoffs, considering both the provider's interests (i.e., revenues) and the customer's interests (i.e., legitimate throughput, response time distribution and costs). Finally, we also assess the cost-effectiveness of our framework under highly variable application service times.
ACM Transactions on Intelligent Systems and Technology (TIST), 2014
Recommender systems are quickly becoming ubiquitous in applications such as e-commerce, social me... more Recommender systems are quickly becoming ubiquitous in applications such as e-commerce, social media channels, and content providers, among others, acting as an enabling mechanism designed to overcome the information overload problem by improving browsing and consumption experience. A typical task in many recommender systems is to output a ranked list of items, so that items placed higher in the rank are more likely to be interesting to the users. Interestingness measures include how accurate, novel, and diverse are the suggested items, and the objective is usually to produce ranked lists optimizing one of these measures. Suggesting items that are simultaneously accurate, novel, and diverse is much more challenging, since this may lead to a conflicting-objective problem, in which the attempt to improve a measure further may result in worsening other measures. In this article, we propose new approaches for multiobjective recommender systems based on the concept of Pareto efficiency—a...
Na construcao de uma aplicacao de aprendizado de maquina, um especialista define seu objetivo, de... more Na construcao de uma aplicacao de aprendizado de maquina, um especialista define seu objetivo, determina, baseado em algumas hipoteses, os dados que possuem uma relacao causal com o objetivo, seleciona o modelo que melhor se adequa a suas hipoteses e dados, realiza experimentos e analisa a qualidade da solucao. Na fase de analise, uma propriedade fundamental do modelo e a interpretabilidade. Em alguns dominios de aplicacao, como medica ou de negocios, a interpretabilidade e tida como diferencial. Para que um modelo seja interpretavel e indicado que ele possua poucos atributos e siga o principio da parcimonia no qual, dentre explicacoes equivalentes, as mais simples sao preferiveis. Esse principio tem se mostrado adequado para classificadores baseados em regras de associacao, nos quais a quantidade de regras utilizadas podem ser substancialmente reduzidas utilizando-se representacoes condensadas, como os conjuntos maximos ou fechados. Porem, a quantidade restante de regras ainda e gr...
The Occam's Razor principle has become the basis for many Machine Learning algorithms, under ... more The Occam's Razor principle has become the basis for many Machine Learning algorithms, under the interpretation that the classifier should not be more complex than necessary. Recently, this principle has shown to be well suited to associative classifiers, where the number of rules composing the classifier can be substantially reduced by using condensed representations such as maximal or closed rules. While it is shown that such a decrease in the complexity of the classifier (usually) does not compromise its accuracy, the number of remaining rules is still larger than necessary and making it hard for experts to interpret the corresponding classifier. In this paper we propose a much more aggressive filtering strategy, which decreases the number of rules within the classifier dramatically without hurting its accuracy. Our strategy consists in evaluating each rule under different statistical criteria, and filtering only those rules that show a positive balance between all the criter...
Performance modeling is a central task in the capacity management of server platforms. Traditiona... more Performance modeling is a central task in the capacity management of server platforms. Traditional performance models are created for transactional workloads (purely open models), batch or interactive (purely closed models). However, many real Web applications experience session-based workloads that exhibit a partially-open behavior, which includes components from either open or closed model, particularly with respect to response time. In this paper, we developed an analytical model for performance that captures the main aspects of partially-open workloads of multi-layered Web platforms and compared it with the open and closed models by simulation. Resumo. A modelagem de desempenhoé uma tarefa central para o gerenciamento de capacidade de plataformas servidoras. Modelos de desempenho tradicionais são derivados para cargas de trabalho transacionais (modelos puramente abertos), batch ou interativas (modelos puramente fechados). Entretanto, várias aplicações reais da Web consistem em cargas baseadas em sessões com comportamento semi-aberto que incluem componentes dos modelos abertos e fechados, particularmente em relação ao tempo de resposta. Neste artigo, desenvolvemos um modelo analítico de desempenho que captura aspectos de cargas semi-abertas para plataformas Web multi-camadas e analisamos seus compromissos em relação aos modelos fechado e aberto por meio de simulação.
In this work we present an extension of our previous capacity management framework making it robu... more In this work we present an extension of our previous capacity management framework making it robust to security attacks. In a scenario where multiple applications are hosted in a shared infrastructure, our performance and optimization models are extended to minimize the impact of security attacks on the provider's revenue and applications' QoS. A number of scenarios and strategies based on dynamic SLA contracts are designed to help uncover the main tradeoffs considering both the provider's interests and the customer's interests.
Capacity management of a hosting infrastructure has traditionally focused only on performance goa... more Capacity management of a hosting infrastructure has traditionally focused only on performance goals. However, the quality of service provided to the hosted applications and the revenues achieved by the provider, depend also on other aspects, such as security and energy constraints. This paper extends our self-adaptive SLA-driven capacity management solution to capture key performance and cost tradeoffs that arise when operating under security attacks and energy constraints. A number of scenarios and strategies based on dynamic SLA contracts are designed to help uncover the main tradeoffs considering both the provider's interests and the customers's interests.
Capacity management of a hosting infrastructure has traditionally focused only on performance goa... more Capacity management of a hosting infrastructure has traditionally focused only on performance goals. However, the quality of service provided to the hosted applications, and ultimately the revenues achieved by the provider, depend also on other aspects, such as security and energy constraints. This paper extends our self-adaptive SLA-driven capacity management solution to capture, in an unified framework, key performance and cost tradeoffs that arise when operating under security attacks and energy constraints. A number of scenarios and strategies based on dynamic SLA contracts are designed to help uncover, via simulation experiments, the main tradeoffs, considering both the provider's interests (i.e., revenues) and the customer's interests (i.e., legitimate throughput, response time distribution and costs). Finally, we also assess the cost-effectiveness of our framework under highly variable application service times.
Uploads
Papers by Itamar Hata