abstract: Threshold logic has long been studied as a means of achieving higher performance and lower power dissipation, providing improvements by condensing simple logic gates into more complex primitives, effectively reducing gate count, pipeline depth, and number of interconnects. This work proposes a new physical implementation of threshold logic, the threshold logic latch (TLL), which overcomes the difficulties observed in previous work, particularly with respect to gate reliability in the presence of noise and process variations. Simple but effective models were created to assess the delay, power, and noise margin of TLL gates for the purpose of determining the physical parameters and assignment of input signals that achieves the lowest delay subject to constraints on power and reliability. From these models, an optimized library of standard TLL cells was developed to supplement a commercial library of static CMOS gates. The new cells were then demonstrated on a number of automatically synthesized, placed, and routed designs. A two-stage 2's complement integer multiplier designed with CMOS and TLL gates utilized 19.5% less area, 28.0% less active power, and 61.5% less leakage power than an equivalent design with the same performance using only static CMOS gates. Additionally, a two-stage 32-instruction 4-way issue queue designed with CMOS and TLL gates utilized 30.6% less area, 31.0% less active power, and 58.9% less leakage power than an equivalent design with the same performance using only static CMOS gates. / Dissertation/Thesis / Ph.D. Computer Science 2010

abstract: It is estimated that wind induced soil transports more than 500 x 106 metric tons of fugitive dust annually. Soil erosion has negative effects on human health, the productivity of farms, and the quality of surface waters. A variety of different polymer stabilizers are available on the market for fugitive dust control. Most of these polymer stabilizers are expensive synthetic polymer products. Their adverse effects and expense usually limits their use. Biopolymers provide a potential alternative to synthetic polymers. They can provide dust abatement by encapsulating soil particles and creating a binding network throughout the treated area. This research into the effectiveness of biopolymers for fugitive dust control involved three phases. Phase I included proof of concept tests. Phase II included carrying out the tests in a wind tunnel. Phase III consisted of conducting the experiments in the field. Proof of concept tests showed that biopolymers have the potential to reduce soil erosion and fugitive dust transport. Wind tunnel tests on two candidate biopolymers, xanthan and chitosan, showed that there is a proportional relationship between biopolymer application rates and threshold wind velocities. The wind tunnel tests also showed that xanthan gum is more successful in the field than chitosan. The field tests showed that xanthan gum was effective at controlling soil erosion. However, the chitosan field data was inconsistent with the xanthan data and field data on bare soil. / Dissertation/Thesis / Ph.D. Civil and Environmental Engineering 2011

abstract: Process variations have become increasingly important for scaled technologies starting at 45nm. The increased variations are primarily due to random dopant fluctuations, line-edge roughness and oxide thickness fluctuation. These variations greatly impact all aspects of circuit performance and pose a grand challenge to future robust IC design. To improve robustness, efficient methodology is required that considers effect of variations in the design flow. Analyzing timing variability of complex circuits with HSPICE simulations is very time consuming. This thesis proposes an analytical model to predict variability in CMOS circuits that is quick and accurate. There are several analytical models to estimate nominal delay performance but very little work has been done to accurately model delay variability. The proposed model is comprehensive and estimates nominal delay and variability as a function of transistor width, load capacitance and transition time. First, models are developed for library gates and the accuracy of the models is verified with HSPICE simulations for 45nm and 32nm technology nodes. The difference between predicted and simulated σ/μ for the library gates is less than 1%. Next, the accuracy of the model for nominal delay is verified for larger circuits including ISCAS'85 benchmark circuits. The model predicted results are within 4% error of HSPICE simulated results and take a small fraction of the time, for 45nm technology. Delay variability is analyzed for various paths and it is observed that non-critical paths can become critical because of Vth variation. Variability on shortest paths show that rate of hold violations increase enormously with increasing Vth variation. / Dissertation/Thesis / M.S. Electrical Engineering 2011

Teste intradérmico avalia reação de hipersensibilidade a diversos agentes que possam apresentar poder reacional alérgico e são comumente utilizados para complementar o diagnóstico da dermatite atópica canina (DAC). Ainda não existe consenso sobre as concentrações de histamina e extratos alergênicos a serem utilizadas. Para determinar a concentração ideal de histamina, como controle positivo, e do limiar irritativo de extratos alergênicos em teste intradérmico é necessário que diversas concentrações sejam avaliadas em uma população bem numerosa de cães hígidos. O objetivo desta pesquisa foi avaliar em 160 cães sadios submetidos a teste intradérmico, quais seriam as concentrações de histamina e de extratos alergênicos consideradas ideais. A solução contendo 0,1 mg/mL de histamina foi considerada como parâmetro ideal, provocando reações cutâneas com diâmetro médio, mediana e desvio padrão, de 15,18 mm, 14,97 mm e 2,07 mm, respectivamente. A partir do estabelecimento da concentração de histamina, foram determinadas as concentrações ótimas dos extratos alergênicos, expressas em PNU/mL: 1.000 para Dermatophagoides pteronyssinus, 500 para D. farinae, 125 para Blomia tropicalis e 2.000 para Malassezia pachydermatis. Futuros estudos devem ser conduzidos em cães atópicos para verificação da acurácia dos testes intradérmicos realizados com essas concentrações. / Intradermal testing evaluates hipersensitivity reaction to different agents that can present allergic reactivity power. It is commonly used to complement canine atopic dermatitis diagnosis. There is still no consensus about histamine concentrations and allergen extracts to be used. The determination of the histamine ideal concentration as positive control and the irritant threshold of allergen extracts for intradermal testing, requires evaluation of different concentrations on a large population of healthy dogs. The purpose of this research was to evaluate the ideal histamine and allergen extracts concentrations on 160 healthy dogs submitted to intradermal testing. A histamine solution 0,1 mg/mL was considered the ideal parameter. It caused cutaneous reactions with average diameter, median measure and standard deviation of 15.18 mm, 14.97 mm and 2.07 mm, respectively. From the histamine concentration establishment, the optimum allergen extracts concentrations were determined, expressed by PNU/mL: 1.000 for Dermatophagoides pteronyssinus, 500 for D. farinae, 125 for Blomia tropicalis and 2.000 for Malassezia pachydermatis. Future studies have to be conducted on atopic dogs to verify the accuracy of the intradermal testing with these concentrations.

Sir Alec Jeffreys was the first to describe human identification with deoxyribonucleic acid (DNA) in his seminal work in 1985 (1); the result was the birth of forensic DNA analysis. Since then, DNA has become the primary substance used to conduct human identification testing. Forensic DNA analysis has evolved since the work of Jeffreys and now incorporates the analysis of 15 to 24 STR (short tandem repeat) locations, or loci (2-4). The simultaneous amplification and subsequent electrophoresis of tens of STR polymorphisms results in analysis that are highly discriminating. DNA target masses of 0.5 to 2 nanograms (ng) are sufficient to obtain a full STR profile (4); however, pertinent information can still be obtained if low copy numbers of DNA are collected from the crime scene or evidentiary material (4-9). Despite the sensitivity of polymerase chain reaction (PCR) - capillary electrophoresis (CE) based technology, low copy DNA signal can be difficult to interpret due to the preponderance of low signal-to-noise ratios. Due to the complicated nature of low template signal, optimization of the DNA laboratory process such that high-fidelity signal is regularly produced is necessary; studies designed to effectively hone in on optimized laboratory conditions are presented herein. The STR regions of a set of samples containing 0.0078 ng of DNA were amplified for 29 cycles; the amplified fragments were separated using two types of CE platforms: an ABI 3130 Genetic Analyzer and an ABI 3500 Genetic Analyzer. The result is a genetic trace, or electropherogram (EPG), comprised of three signal components that include noise, artifact, and allele. The EPGs were analyzed using two peak detection software programs. In addition, a tool, termed Simulating Evidentiary Electropherograms (SEEIt) (10, 11), was utilized to simulate EPG signal obtained when one copy of DNA is processed through the forensic pipeline. SEEIt was parameterized to simulate data corresponding to two laboratory scenarios: the amplification of a single copy of DNA injected on an ABI 3130 Genetic Analyzer and on an ABI 3500 Genetic Analyzer. In total, 20,000 allele peaks and 20,000 noise peaks were generated for each CE platform. Comparison of simulated and experimental data was used to elucidate features that are difficult to ascertain by experimental work alone. The data demonstrate that experimental signal obtained with the ABI 3500 platform results in signal that is, on average, a factor of four larger than signal obtained from the ABI 3130 platform. When a histogram of the signal is plotted, a multi modal distribution is observed. The first mode is hypothesized to be the result of noise, while the second, third, etc. modes are the signal obtained when one, two, etc. target DNA molecules are amplified. By evaluating the data in this way, full signal resolution between noise and allelic signal is visualized. Therefore, this methodology may be used to: 1) optimize post-PCR laboratory conditions to obtain excellent resolution between noise and allelic signal; and 2) determine an analytical threshold (AT) that results in few false detections and few cases of allelic dropout. A χ2 test for independence of the experimental signal in noise positions and the experimental signal within allele positions < 12 relative fluorescence units (RFU), i.e. signal in the noise regime, indicate the populations are not independent when sufficient signal-to-noise resolution is obtained. Once sufficient resolution is achieved, optimized ATs may be acquired by evaluating and minimizing the false negative and false positive detection rates. Here, a false negative is defined as the non-detection of an allele and a false positive is defined as the detection of noise. An AT of 15 RFU was found to be the optimal AT for samples injected on the ABI 3130 for at least 10 seconds (sec) as 99.42% of noise peaks did not exceed this critical value while allelic dropout was kept to a minimum, 36.97%, at this AT. Similarily, in examining signal obtained from the ABI 3500, 99.41% and 99.0% of noise fell under an AT of 50 RFU for data analyzed with GeneMapper ID-X (GM) and OSIRIS (OS), respectively. Allelic dropout was 36.34% and 36.55% for GM and OS, respectively, at this AT.

abstract: Threshold logic has been studied by at least two independent group of researchers. One group of researchers studied threshold logic with the intention of building threshold logic circuits. The earliest research to this end was done in the 1960's. The major work at that time focused on studying mathematical properties of threshold logic as no efficient circuit implementations of threshold logic were available. Recently many post-CMOS (Complimentary Metal Oxide Semiconductor) technologies that implement threshold logic have been proposed along with efficient CMOS implementations. This has renewed the effort to develop efficient threshold logic design automation techniques. This work contributes to this ongoing effort. Another group studying threshold logic did so, because the building block of neural networks - the Perceptron, is identical to the threshold element implementing a threshold function. Neural networks are used for various purposes as data classifiers. This work contributes tangentially to this field by proposing new methods and techniques to study and analyze functions implemented by a Perceptron After completion of the Human Genome Project, it has become evident that most biological phenomenon is not caused by the action of single genes, but due to the complex interaction involving a system of genes. In recent times, the `systems approach' for the study of gene systems is gaining popularity. Many different theories from mathematics and computer science has been used for this purpose. Among the systems approaches, the Boolean logic gene model has emerged as the current most popular discrete gene model. This work proposes a new gene model based on threshold logic functions (which are a subset of Boolean logic functions). The biological relevance and utility of this model is argued illustrated by using it to model different in-vivo as well as in-silico gene systems. / Dissertation/Thesis / Ph.D. Computer Science 2012

Resumo: Várias características são avaliadas por meio de escores visuais nos programas de avaliação genética no Brasil. Os sistemas de atribuição dos escores variam quanto ao número de classes de notas, ao referencial adotado para atribuição das notas e à frequência das notas nas diferentes classes e ao modelo empregado na análise dos dados. O objetivo neste estudo foi avaliar o impacto de diferentes escalas de atribuição dos escores, modelos de análise dos dados e sua interação na identificação dos genótipos superiores para essas características, por meio de dados simulados. Foram simulados dados de uma característica contínua na escala subjacente, transformados em valores (1 a 6) na escala discreta observada, com diferentes escalas (relativa uniforme, relativa normal, relativa assimétrica e fixa), dois valores de herdabilidade (0,25 e 0,49) e duas variâncias de grupos de contemporâneos (0,25 e 1,00). Estimaram-se os valores genéticos utilizando modelos de limiar, linear e linear após transformação dos dados. Análises com modelo de limiar e notas com escala relativa normal apresentaram melhores acurácias para a predição dos valores genéticos. A utilização do modelo de limiar incrementa a acurácia das estimativas, mas em razão da demanda computacional e de problemas de implementação, o modelo linear pode ser empregado na análise de dados categóricos avaliados sob escala relativa normal com seis classes. A transformação de dados proposta foi eficiente em algumas situações, podendo ser considerada como alternativa ao modelo linear, dependendo da estrutura dos dados / Abstract: Genetic improvement programs in Brazil involve the monitoring of many traits that are evaluated by means of visual scores. The systems used to evaluate animals vary in terms of the number of classes, the referential from which the scores are given, the distribution of observations in score classes and the model used to analyze the data. The objective of this study was to evaluate the impact of different scales of score evaluation, models of analysis and their interactions on the identification of superior genotypes for visual scores, by means of a simulation study. Data of a continuous trait were simulated in a subjacent scale and then transformed into a discrete trait (1 to 6) in the observable scale, with different scale distributions (uniform relative, normal relative, asymmetric relative and fixed), two heritabilities (0.25 and 0.49) and two contemporary group variances (0.25 and 1.00). Breeding values were estimated using threshold model, linear model and linear model after data transformation. The threshold model and the relative normal scale showed better accuracy for the prediction of breeding values. The use of a threshold model can lead to an increase in the estimates' accuracy, but despite the computational demand and implementation issues, the linear model can be used to analyze discrete data assigned under a normal relative scale with six score classes. The proposed data transformation was effective for some situations and depending on the data structure it should be considered as an alternative option instead of linear model / Orientador: Maurício Mello de Alencar / Coorientador: Roberto Augusto de Almeida Torres Junior / Banca: Gerson Barreto Mourão / Banca: Elias Nunes Martins / Banca: Danísio Prado Munari / Banca: Henrique Nunes de Oliveira / Doutor

This diploma thesis is focused on testing drug users. Specifically, this thesis is concerned with testing drug users in low-threshold centres.These low-threshold centres for drug users operate in the area of secondary and tertiary prevention. Furthermore, it describes individual diseases such as HIV, HBV, HCV and syphilis, their mode of transmission, clinical course, diagnosis, treatment and possible prevention, drug policy of the Czech Republic, the conditions for testing the abovementioned diseases in low-threshold centres and the legal framework for testing at these facilities.The practical part is divided into several research studies.The qualitative research is aimed at low-threshold centres. In this part, I am trying to determine the number of HIV, HBV and HCV tests performed in 2014. Next, the purpose of this section is also to define where the testing takes place and who is tested. The research data were obtained using a questionnaire distributed to the low-threshold centres in the CR. According to the Annual Report on Drug Epidemiology from 2014, there are 76 low-threshold centres in the Czech Rep. A total of 32 questionnaires were collected. The questionnaire consisted of 32 questions. Close-ended questions were used to detect the type of facility, whether the facility carries out tests for the chosen types of diseases, what type of test is used, who performs the testing and where the testing is carried out. Open-ended questions were used to determine the number of people tested. The qualitative research consists of 18 interviews with drug users from the Contact Centre Český Krumlov. Thanks to its field program, this centre operates in Český Krumlov, Větřní, Kaplice, Horní Planá, Loučovice, Frymburk, Vyšší Brod and in border regions such as Horní and Dolní Dvořiště or Studánky where the drug abuse is very problematic. The data were collected using a semistructured anonymous interview. The interview consisted of 4 areas:Data for identification purposes. Drug history.Testing.Awareness.30 questionnaires received from the respondents at the Prevent Contact Centre in České Budějovice can be found in the Appendix. The data were collected using an anonymous questionnaire which consisted of three basic areas. The collected data included drug history and identification information, information about testing for infectious diseases and knowledge of health risks. The thesis states 6 goals. The first 3 goals deal with mapping of HBV, HCV and HIV testing among drug users at low-threshold centres in 2014. The purpose of the fourth goal is to determine how the testing is carried out in low-threshold centres. The fifth goal attempts to establish whether drug users are interested in HBV, HCV and HIV testing. Finally, the sixth goal determines whether drug users were tested in 2014.After processing the results I established the following hypotheses which resulted from the research.H1: Drug users are interested in their health.H2:Drug users are aware of risks that result from these diseases.H3: The possibility of testing at low-threshold centres is not chosen very often by drug users.The research showed that drug users have adequate information about the risks that are associated with drug injection and injecting equipment sharing, but the average amount of performed tests is relatively low.To sum up, drug users need to be constantly informed about the importance of regular testing. In order to increase the number of tested people, it would be advisable to adapt the testing to the users' needs and also to perform more tests within field programs. It would also be helpful to link health facilities to contact centres locally.

Cenário clínico: A síndrome dolorosa miofascial (SDM), causa comum de dor musculoesquelética, pode ser incapacitante e desafiadora terapeuticamente, devido à ineficácia dos tratamentos convencionais para dor. Intervenções terapêuticas alternativas precisam ser pesquisadas para alcançar vias do processo de doença não contempladas com a terapêutica clássica. Dentre estas, o uso da melatonina, com efeitos cronobiótico, ansiolítico e analgésico, tem se apresentado como uma opção terapêutica atrativa no tratamento da SDM, que cursa com alterações de sono, dor, sintomas depressivos e de ansiedade. Objetivos: Avaliar a eficácia da melatonina exógena na redução da dor, no limiar de dor à pressão (LDP) e na qualidade de sono de pacientes com SDM facial. Métodos e Resultados: Um estudo randomizado, controlado foi realizado em 45 mulheres com dor miofascial, com idades entre 18 e 40 anos, segundo critérios Research Diagnostic Criteria for Temporomandibular Disorder (RDC/TMD). A eficácia da melatonina oral foi avaliada na redução da dor e melhora tanto do limiar de dor a pressão (LDP) como da qualidade do sono. Os participantes foram randomizados para receber 5 mg / dia de melatonina, 5 mg / dia ciclobenzaprina, ou placebo durante um período de quatro semanas. O efeito absoluto das intervenções, apresentado como ES (tamanho do efeito) sobre a dor: placebo versus melatonina foi de 2,08 (1,17-2,97) e de ciclobenzaprina vs placebo foi de -1,25 (0,45-2,06)]. O número de pacientes necessários para tratar (NNT) para evitar a dor moderada a intensa foi 3 (95% CI, 2-4) e 18 (95% IC, 9 a a) nos grupos de melatonina e de ciclobenzaprina, respectivamente, em relação ao placebo. O ES no LDP melatonina vs placebo e ciclobenzaprina vs placebo foi de 2,72 (1,69-3,75) e 1,01 (0,23-1,79), respectivamente. O ES na escala visual analógica de Qualidade de Sono (VASQS) utilizada para avaliar a forma como as pacientes se sentiram ao acordar, durante o período de tratamento, foi nos grupos melatonina versus placebo de 2,47 (1,49-3,45) e 1,01 (0,23-1,79), respectivamente. Conclusão: Melatonina foi mais eficaz do que placebo para melhorar a dor miofascial crônica facial e ambos os tratamentos foram mais eficazes do que placebo para melhorar o LDP e a qualidade de sono. / Background: The Myofascial Pain Syndrome (SDM), a common cause of musculoskeletal pain, can course with disability and can be a therapeutical challenge, due to the ineffectiveness of conventional treatments for pain. Alternative therapeutic interventions must be researched to achieve the process of the disease process that in not dealt with the classical therapy. Among these, the use of melatonin, which takes effect chronobiotic, anxiolytic and analgesic, has been presented as an attractive therapeutic option in the treatment of SDM, which leads to sleep disturbances, pain, anxiety and depressive symptoms. Objectives: Evaluate the efficacy of exogenous melatonin in reducing pain, pain pressure threshold (PPT) and the sleep quality of patients with chronic myofascial face pain. Methods and Results: A randomized, controlled trial was conducted with 45 females, aged 18 to 40 years who presented myofascial pain according to the Research Diagnostic Criteria for Temporomandibular Disorder (RDC/TMD) guidelines. The efficacy of oral melatonin was evaluated in reducing pain and improving both the pain pressure threshold (PPT) and sleep quality. Participants were randomized to receive 5 mg/day melatonin, 5 mg/day cyclobenzaprine, or a placebo during a four-week period. The absolute effect of interventions, presented as ES (effect size) on pain for melatonin vs. placebo was 2.08 (1.17 to 2.97) and for cyclobenzaprine vs. placebo -1.25 (0.45 to 2.06)], respectively. The Number of Patients Needed to be Treated (NNT) to prevent moderate to intense pain was 3 (95% CI, 2 to 4) and 18 (95% CI, 9 to ) in the melatonin and cyclobenzaprine groups, respectively compared to the placebo. The ES on the PPT for melatonin vs. placebo and cyclobenzaprine vs. placebo was 2.72 (1.69 to 3.75) and 1.01 (0.23 to 1.79), respectively. The ES on the Visual Analog Sleep Quality Scale (VASQS) scores used to assess how they felt when they woke up during the treatment period for the melatonin vs. placebo were 2.47 (1.49 to 3.45) and 1.01 (0.23 to 1.79), respectively. Conclusion: Melatonin was more effective than placebo for improving chronic myofascial face pain and both treatments were more effective than placebo for improving sleep quality and the PPT.

Em sistemas reais, os dados armazenados tipicamente apresentam inconsistências causadas por erros de gra a, abreviações, caracteres trocados, entre outros. Isto faz com que diferentes representações do mesmo objeto do mundo real sejam registrados como elementos distintos, causando um problema no momento de consultar os dados. Portanto, o problema investigado nesta tese refere-se às consultas por abrangência, que procuram encontrar objetos que representam o mesmo objeto real consultado . Esse tipo de consulta não pode ser processado por coincidência exata, necessitando de um mecanismo de consulta com suporte à similaridade. Para cada consulta submetida a uma determinada coleção, a função de similaridade produz um ranking dos elementos dessa coleção ordenados pelo valor de similaridade entre cada elemento e o objeto consulta. Como somente os elementos que são variações do objeto consulta são relevantes e deveriam ser retornados, é necessário o uso de um limiar para delimitar o resultado. O primeiro desa o das consultas por abrangência é a de nição do limiar. Geralmente é o especialista humano que faz a estimativa manualmente através da identi - cação de elementos relevantes e irrelevantes para cada consulta e em seguida, utiliza uma medida como revocação e precisão (R&P). A alta dependência do especialista humano di culta o uso de consultas por abrangência na prática, principalmente em grandes coleções. Por esta razão, o método apresentado nesta tese tem por objetivo estimar R&P para vários limiares com baixa dependência do especialista humano. Como um sub-produto do método, também é possível selecionar o limiar mais adequado para uma função sobre uma determinada coleção. Considerando que as funções de similaridade são imperfeitas e que apresentam níveis diferentes de qualidade, é necessário avaliar a função de similaridade para cada coleção, pois o resultado é dependente dos dados. Um limiar para uma coleção pode ser totalmente inadequado para outra coleção, embora utilizando a mesma função de similaridade. Como forma de medir a qualidade de funções de similaridade no contexto de consultas por abrangência, esta tese apresenta a discernibilidade. Trata-se de uma medida que de ne a habilidade da função de similaridade de separar elementos relevantes e irrelevantes. Comparando com a precisão média, a discernibilidade captura variações que não são percebidas pela precisão média, o que mostra que a discernibilidade é mais apropriada para consultas por abrangência. Uma extensa avaliação experimental usando dados reais mostra a viabilidade tanto do método de estimativas como da medida de discernibilidade para consultas por abrangência. / In real systems, stored data typically have inconsistencies caused by typing errors, abbreviations, transposed characters, amongst others. For this reason, di erent representations of the same real world object are stored as distinct elements, causing problems during query processing. In this sense, this thesis investigates range queries which nd objects that represent the same real world object being queried . This type of query cannot be processed by exact matching, thus requiring the support for querying by similarity. For each query submitted to a given collection, the similarity function produces a ranked list of all elements in this collection. This ranked list is sorted decreasingly by the similarity score value with the query object. Only the variations of the query object should be part of the result as only those items are relevant. For this reason, it is necessary to apply a threshold value to properly split the ranking. The rst challenge of range queries is the de nition of a proper threshold. Usually, a human specialist makes the estimation manually through the identi cation of relevant and irrelevant elements for each query. Then, he/she uses measures such as recall and precision (R&P). The high dependency on the human specialist is the main di culty related to use of range queries in real situations, specially for large collections. In this sense, the method presented in this thesis has the objective of estimating R&P at several thresholds with low human intervention. As a by-product of this method, it is possible to select the optimal threshold for a similarity function in a given collection. Considering the fact that the similarity functions are imperfect and vary in quality, it is necessary to evaluate the similarity function for each collection as the result is domain dependent. A threshold value for a collection could be totally inappropriate for another, even though the same similarity function is applied. As a measure of quality of similarity functions for range queries, this thesis introduces discernability. This is a measure to quantify the ability of the similarity function in separating relevant and irrelevant elements. Comparing discernability and mean average precision, the rst one can capture variations that are not noticed by precision-based measures. This property shows that discernability presents better results for evaluating similarity functions for range queries. An extended experimental evaluation using real data shows the viability of both, the estimation method and the discernability measure, applied to range queries.