No description available.

Geralmente, o resultado de um exame diagnóstico é transmitido através de um relatório, elaborado pelo médico que o executa para o médico solicitante. A elaboração livre, sem qualquer regra ou padrão, do laudo pelo médico executante, é rotineira, porém gera reclamações por parte dos médicos solicitantes, freqüentemente relacionadas à terminologia e ao conteúdo, que nem sempre é completo, preciso e conclusivo. O objetivo desse trabalho foi estudar a viabilidade, identificando os obstáculos, para implantação de um laudo estruturado, em um hospital universitário e, se isso resulta em melhoria na transmissão de informações. O laudo escolhido foi o laudo de ultra-sonografia de tireóide. As informações e terminologias foram padronizadas, utilizando um software de laudo estruturado, construído a partir de informações dos médicos solicitantes e executantes, em ambiente de internet. A ferramenta utilizada para construção do software foi o Borland Developer Studio 2006 (Zarko, 2009), sob plataforma Microsoft.Net (Microsoft, 2009). Durante dois meses, o tempo para elaboração do laudo textual (livre) foi medido e gravado. O mesmo foi feito com o laudo estruturado nos dois meses seguintes. Ambos os modelos de laudo foram avaliados, após o período de utilização, através de questionários específicos, aplicados aos médicos solicitantes e executantes para avaliar a mudança do textual para o estruturado. Cento e cinqüenta e sete (157) exames com tempo médio igual a 7,1 (± 4,5) minutos foram elaborados sob a forma livre. Com o uso do laudo estruturado, 100 exames com tempo médio igual a 5,4 (± 3,6) minutos. A diferença do tempo total (laudos com presença de nódulos e sem presença de nódulos) e do tempo com nódulos entre os laudos gerados estruturado e os por texto livre foi significativa após a análise estatística (p-valor < 0,01). No entanto, quando se considera apenas os laudos sem nódulos,a diferença não é significativa (p-valor=0,20). Entre os médicos solicitantes (n=10), a maioria (8/10 = 80%) preferiu o laudo estruturado. Todos concordaram que houve padronização da descrição e 70%, que houve melhora na transmissão de informação. Dos médicos executantes (n=21), 16 (76,2%) também preferiram o laudo estruturado porque padronizou a descrição dos achados (20/21 = 95,2%) e facilitou a emissão do laudo (14/21 = 66,7%). Este trabalho concluiu que é viável a implantação de um laudo estruturado em um hospital universitário, melhorando a velocidade da elaboração do laudo e a transmissão das informações entre médico solicitante e executante. / Usually, the findings of an imaging study are transmitted by a written report, from the radiologist to the requiring physician. The free text (FT), without precise and defined rules, is generally applied for this purpose. However, results in complaints by clinicians, mainly related to imprecise terms, vague and inconclusive content. The purpose of this study was to verify the feasibility of making structured report (SR), through the identification of the obstacles, within university hospital and the impact over information transmission. Thyroid ultrasonography (Thyroid US) was the exam chosen. The information and terms was standardized, and the SR was made based on information from clinicians and radiologists using web-based software, the Borland Developer Studio 2006, under Microsoft Net. During two months, time used to generate a report for Thyroid US was measured using FT. The same was done in the next two months with SR. Both models of reports were evaluated by questionnaire directed to clinicians and radiologists. 157 exams, average time 7.1 +/- 4.5 minutes were generated using FT and 100, average time 5.4+/-3.6 using SR. The difference of time to report between the SR and the FT was significant after the statistical analysis (p-value <0,01).for all examinations (reports with and without nodules) and for examinations with nodules However, when is just considered the reports without nodules, the difference was not significant (p-valor=0,20). Among the clinicians, the majority (80%- 8/10) preferred the SR. All of them agreed that standardization was reached and 70% stated that improved information transmission. 76.2%, sixteen out of twenty-one, of the radiologists preferred the SR because the description was more uniform and facilitates report generation. In conclusion, this study showed that implementation of SR in an university hospital is feasible and improves information transmission between clinicians and radiologists.

Introdução: O laudo radiológico com entrada de dados estruturada proporciona precisão na recuperação de informações, todavia, é menos abrangente que o texto-livre e pode impedir o especialista de descrever um achado importante, não contemplado pela estrutura pré-definida. Portanto, a solução está em interpretar o texto escrito pelo especialista e os transportar para uma estrutura computacional ontológica formalizando o conhecimento. Objetivos: Elaborar e avaliar uma metodologia para estruturação de informações de laudos radiológicos em texto livre que contemple as especificidades brasileiras, incluindo a terminologia empregada. Métodos: Coleta de 1701 laudos radiológicos de ressonância magnética da coluna lombossacra em três instituições distintas. Para estruturar a informação, foram utilizadas técnicas de mineração de texto e conceituação ontológica das unidades lexicais. Dez radiologistas especialistas avaliaram a superestrutura textual e a terminologia extraída e conceituada, por meio de questionário eletrônico onde foram sugeridas 3 superestruturas textuais e 30 termos com a descrição do respectivo conceito. Resultados: A metodologia criada é constituída por 6 passos: 1) coleta dos laudos radiológicos de determinado tipo de exame; 2) decomposição textual; 3) normalização das unidades lexicais; 4) identificação das superestruturas textuais; 5) conceituação dos termos candidatos; 6) avaliação das superestruturas identificadas e da terminologia extraída e conceituada pelos especialistas na linguagem de especialidade estudada. Foram identificadas três superestruturas textuais diferentes. Após processamento automático inicial, 6129 unidades lexicais foram normalizadas obtendo 4568 termos candidatos dos quais 4183 foram conceituados e originaram 727 conceitos. Foram estabelecidas 13963 relações entre termos candidatos e conceitos, e 789 relações entre conceitos. Quanto a adequada descrição dos termos, 40% dos avaliadores concordam plenamente que estão adequados, 60% concordam parcialmente que estão adequados e nenhum avaliador discordou que os termos são adequados. Em relação a quantidade de termos utilizados pelos avaliadores, em sua respectiva instituição, 70% utilizam de 24 a 30 termos e 30% de 16 a 23 termos. A conceituação de um termo foi considerada inadequada por dois avaliadores. Conclusão: A metodologia proposta foi avaliada por radiologistas especialistas que identificaram as diferenças das superestruturas textuais entre as instituições e os termos com a descrição do respectivo conceito. Portanto, foi possível estruturar a informação de laudos radiológicos da coluna lombossacra a partir do texto livre identificando as superestruturas textuais, extraindo as unidades lexicais, normalizando as unidades lexicais e conceituando ontologicamente os termos candidatos preservando a referência com as respectivas partes constituintes dos laudos radiológicos produzidos no contexto brasileiro. / Introduction: The radiological report with structured data entry provides precision in information retrieval, however, is less comprehensive than the text-free and may prevent the expert to describe an important finding, not covered by the pre-defined structure. So the solution is to interpret the text written by expert and carry for a computational framework formalizing the ontological knowledge. Objectives: Develop and evaluate a methodology for structuring information in free text radiology reports that contemplates the Brazilian specifications, including terminology. Methods: 1701 radiological reports of lumbosacral MRI were collected in three different institutions. To structure the information we used techniques of text mining and ontological conceptualization of lexical units extracted. Ten dedicated radiologists evaluated the textual superstructure and terminology extracted by the electronic questionnaire which were suggested textual 3 superstructures and 30 terms with the description of its concept. Results: The established methodology consisted of six steps: 1) collection of radiology reports of certain type of examination; 2) textual decomposition; 3) lexical units normalization; 4) textual superstructures identification ; 5) conceptualization of candidate terms; 6) evaluation of identified superstructures and terminology extracted by experts in the specialized language study. We identified three different textual superstructures. After automatic processing initial 6129 lexical units were normalized getting 4568 candidate terms of which 4183 were conceptualized and originated 727 concepts. 13963 were established relations between candidate terms and concepts and 789 relations between concepts. As for adequate description of the terms, 40% of the evaluators agree fully that are appropriate, 60% somewhat agree that they are adequate and no evaluator disagreed that the terms are adequate. Regarding the number of words used by evaluators in their respective institution using 70% between 24 and 30 terms and 30% from 16 to 23 terms. The conceptualization of a term was considered inadequate by two evaluators. The established methodology consisted of six steps: 1) collection of radiology reports of certain type of examination; 2) textual decomposition; 3) lexical units normalization; 4) textual superstructures identification ; 5) conceptualization of candidate terms; 6) evaluation of identified superstructures and terminology extracted by experts in the specialized language study. Conclusions: The proposed methodology was evaluated by radiologists who identified the differences between the textual superstructures institutions and the terms describing the relevant concept. Therefore, it was possible to structure the information of the lumbosacral spine radiology reports from free text by identifying textual superstructures, extracting lexical units, normalizing the lexical units and conceptualizing the terms ontologically candidates preserving the reference to the respective constituent parts of radiology reports produced in the Brazilian context.

Geralmente, o resultado de um exame diagnóstico é transmitido através de um relatório, elaborado pelo médico que o executa para o médico solicitante. A elaboração livre, sem qualquer regra ou padrão, do laudo pelo médico executante, é rotineira, porém gera reclamações por parte dos médicos solicitantes, freqüentemente relacionadas à terminologia e ao conteúdo, que nem sempre é completo, preciso e conclusivo. O objetivo desse trabalho foi estudar a viabilidade, identificando os obstáculos, para implantação de um laudo estruturado, em um hospital universitário e, se isso resulta em melhoria na transmissão de informações. O laudo escolhido foi o laudo de ultra-sonografia de tireóide. As informações e terminologias foram padronizadas, utilizando um software de laudo estruturado, construído a partir de informações dos médicos solicitantes e executantes, em ambiente de internet. A ferramenta utilizada para construção do software foi o Borland Developer Studio 2006 (Zarko, 2009), sob plataforma Microsoft.Net (Microsoft, 2009). Durante dois meses, o tempo para elaboração do laudo textual (livre) foi medido e gravado. O mesmo foi feito com o laudo estruturado nos dois meses seguintes. Ambos os modelos de laudo foram avaliados, após o período de utilização, através de questionários específicos, aplicados aos médicos solicitantes e executantes para avaliar a mudança do textual para o estruturado. Cento e cinqüenta e sete (157) exames com tempo médio igual a 7,1 (± 4,5) minutos foram elaborados sob a forma livre. Com o uso do laudo estruturado, 100 exames com tempo médio igual a 5,4 (± 3,6) minutos. A diferença do tempo total (laudos com presença de nódulos e sem presença de nódulos) e do tempo com nódulos entre os laudos gerados estruturado e os por texto livre foi significativa após a análise estatística (p-valor < 0,01). No entanto, quando se considera apenas os laudos sem nódulos,a diferença não é significativa (p-valor=0,20). Entre os médicos solicitantes (n=10), a maioria (8/10 = 80%) preferiu o laudo estruturado. Todos concordaram que houve padronização da descrição e 70%, que houve melhora na transmissão de informação. Dos médicos executantes (n=21), 16 (76,2%) também preferiram o laudo estruturado porque padronizou a descrição dos achados (20/21 = 95,2%) e facilitou a emissão do laudo (14/21 = 66,7%). Este trabalho concluiu que é viável a implantação de um laudo estruturado em um hospital universitário, melhorando a velocidade da elaboração do laudo e a transmissão das informações entre médico solicitante e executante. / Usually, the findings of an imaging study are transmitted by a written report, from the radiologist to the requiring physician. The free text (FT), without precise and defined rules, is generally applied for this purpose. However, results in complaints by clinicians, mainly related to imprecise terms, vague and inconclusive content. The purpose of this study was to verify the feasibility of making structured report (SR), through the identification of the obstacles, within university hospital and the impact over information transmission. Thyroid ultrasonography (Thyroid US) was the exam chosen. The information and terms was standardized, and the SR was made based on information from clinicians and radiologists using web-based software, the Borland Developer Studio 2006, under Microsoft Net. During two months, time used to generate a report for Thyroid US was measured using FT. The same was done in the next two months with SR. Both models of reports were evaluated by questionnaire directed to clinicians and radiologists. 157 exams, average time 7.1 +/- 4.5 minutes were generated using FT and 100, average time 5.4+/-3.6 using SR. The difference of time to report between the SR and the FT was significant after the statistical analysis (p-value <0,01).for all examinations (reports with and without nodules) and for examinations with nodules However, when is just considered the reports without nodules, the difference was not significant (p-valor=0,20). Among the clinicians, the majority (80%- 8/10) preferred the SR. All of them agreed that standardization was reached and 70% stated that improved information transmission. 76.2%, sixteen out of twenty-one, of the radiologists preferred the SR because the description was more uniform and facilitates report generation. In conclusion, this study showed that implementation of SR in an university hospital is feasible and improves information transmission between clinicians and radiologists.

Introdução: O laudo radiológico com entrada de dados estruturada proporciona precisão na recuperação de informações, todavia, é menos abrangente que o texto-livre e pode impedir o especialista de descrever um achado importante, não contemplado pela estrutura pré-definida. Portanto, a solução está em interpretar o texto escrito pelo especialista e os transportar para uma estrutura computacional ontológica formalizando o conhecimento. Objetivos: Elaborar e avaliar uma metodologia para estruturação de informações de laudos radiológicos em texto livre que contemple as especificidades brasileiras, incluindo a terminologia empregada. Métodos: Coleta de 1701 laudos radiológicos de ressonância magnética da coluna lombossacra em três instituições distintas. Para estruturar a informação, foram utilizadas técnicas de mineração de texto e conceituação ontológica das unidades lexicais. Dez radiologistas especialistas avaliaram a superestrutura textual e a terminologia extraída e conceituada, por meio de questionário eletrônico onde foram sugeridas 3 superestruturas textuais e 30 termos com a descrição do respectivo conceito. Resultados: A metodologia criada é constituída por 6 passos: 1) coleta dos laudos radiológicos de determinado tipo de exame; 2) decomposição textual; 3) normalização das unidades lexicais; 4) identificação das superestruturas textuais; 5) conceituação dos termos candidatos; 6) avaliação das superestruturas identificadas e da terminologia extraída e conceituada pelos especialistas na linguagem de especialidade estudada. Foram identificadas três superestruturas textuais diferentes. Após processamento automático inicial, 6129 unidades lexicais foram normalizadas obtendo 4568 termos candidatos dos quais 4183 foram conceituados e originaram 727 conceitos. Foram estabelecidas 13963 relações entre termos candidatos e conceitos, e 789 relações entre conceitos. Quanto a adequada descrição dos termos, 40% dos avaliadores concordam plenamente que estão adequados, 60% concordam parcialmente que estão adequados e nenhum avaliador discordou que os termos são adequados. Em relação a quantidade de termos utilizados pelos avaliadores, em sua respectiva instituição, 70% utilizam de 24 a 30 termos e 30% de 16 a 23 termos. A conceituação de um termo foi considerada inadequada por dois avaliadores. Conclusão: A metodologia proposta foi avaliada por radiologistas especialistas que identificaram as diferenças das superestruturas textuais entre as instituições e os termos com a descrição do respectivo conceito. Portanto, foi possível estruturar a informação de laudos radiológicos da coluna lombossacra a partir do texto livre identificando as superestruturas textuais, extraindo as unidades lexicais, normalizando as unidades lexicais e conceituando ontologicamente os termos candidatos preservando a referência com as respectivas partes constituintes dos laudos radiológicos produzidos no contexto brasileiro. / Introduction: The radiological report with structured data entry provides precision in information retrieval, however, is less comprehensive than the text-free and may prevent the expert to describe an important finding, not covered by the pre-defined structure. So the solution is to interpret the text written by expert and carry for a computational framework formalizing the ontological knowledge. Objectives: Develop and evaluate a methodology for structuring information in free text radiology reports that contemplates the Brazilian specifications, including terminology. Methods: 1701 radiological reports of lumbosacral MRI were collected in three different institutions. To structure the information we used techniques of text mining and ontological conceptualization of lexical units extracted. Ten dedicated radiologists evaluated the textual superstructure and terminology extracted by the electronic questionnaire which were suggested textual 3 superstructures and 30 terms with the description of its concept. Results: The established methodology consisted of six steps: 1) collection of radiology reports of certain type of examination; 2) textual decomposition; 3) lexical units normalization; 4) textual superstructures identification ; 5) conceptualization of candidate terms; 6) evaluation of identified superstructures and terminology extracted by experts in the specialized language study. We identified three different textual superstructures. After automatic processing initial 6129 lexical units were normalized getting 4568 candidate terms of which 4183 were conceptualized and originated 727 concepts. 13963 were established relations between candidate terms and concepts and 789 relations between concepts. As for adequate description of the terms, 40% of the evaluators agree fully that are appropriate, 60% somewhat agree that they are adequate and no evaluator disagreed that the terms are adequate. Regarding the number of words used by evaluators in their respective institution using 70% between 24 and 30 terms and 30% from 16 to 23 terms. The conceptualization of a term was considered inadequate by two evaluators. The established methodology consisted of six steps: 1) collection of radiology reports of certain type of examination; 2) textual decomposition; 3) lexical units normalization; 4) textual superstructures identification ; 5) conceptualization of candidate terms; 6) evaluation of identified superstructures and terminology extracted by experts in the specialized language study. Conclusions: The proposed methodology was evaluated by radiologists who identified the differences between the textual superstructures institutions and the terms describing the relevant concept. Therefore, it was possible to structure the information of the lumbosacral spine radiology reports from free text by identifying textual superstructures, extracting lexical units, normalizing the lexical units and conceptualizing the terms ontologically candidates preserving the reference to the respective constituent parts of radiology reports produced in the Brazilian context.

The ubiquity and speed of Internet access led over the past decade to an exponential increase in the use of thin clients and cloud computing, both taking advantage of the ability to remotely provide computing resources. The work investigates the role of remote display Protocol in behavioral biometric systems based on free-text keystroke dynamics. Authentication based on keystroke dynamics is easy in use, cheap, invisible for user and does not require any additional sensor.I n this project I will investigate how network characteristics affect the keystroke dynamics pattern in remote desktop scenario. Objectives: The aim of this project is to investigate the role of remote display Protocol in behavioral biometric system based on free-text keystroke dynamics, by measuring how network characteristics influence the computation of keystroke pattern in Virtual Desktop Infrastructure (VDI). Method: This thesis will answer all of its research question with the help of a Systematic Literature Review (SLR) and an Experiment. Literature review was conducted to gather information about the keystroke dynamics analysis, the applied algorithms and their performance; and to clarify the controlled changes of networking performance in VDI based scenario. Using the acquired knowledge, implemented keystroke dynamics pattern algorithm based on Euclidian distance statistical method, designed an experiment and performed a series of tests, in order to identify the influence of remote display protocol to keystroke pattern. Results: Through the SLR, keystroke dynamics analysis working structure is identified and illustrated, essential elements are summarized, and a statistical approach based on Euclidian distance is described; a technique to simulate and measure networklatency in VDI scenario is described including essential elements and parameters of VDI testbed. Keystroke analysis algorithm, dataset replication code and VDItestbed are implemented. The controlled experiment provided measurements of the metrics of the algorithm and network performance mentioned in objectives. Conclusions: During experimentation, I found that timing pattern in the keystroke dynamics data is affected by VDI in normal network conditions by 12% in average. Higher latency standard deviation, jitter, packet loss as well as remote display protocol overheads have a significant combined impact onto keystroke pattern. Moreover I found what maximum possible delay values does not affect keystroke pattern in a larger extent.

Nowadays it is common for companies to sit on and gather a lot of data related to their business. The size of this data is often too large to be analyzed by hand and it is therefore becoming more and more common to automate this analysis e.g. by running machine learning methods on this data. In this project we attempt at analyzing an unstructured dataset consisting of remarks, found by inspectors, on electrical installations. This is done by firstly clustering the dataset with the goal of having each cluster representing a specific type of error found in the field and then extracting ten keywords from each cluster. We investigate whether these keywords can be used for representing the clusters’ contents in a way that could be useful for a future end-user application. The solution developed in this project was evaluated by constructing a form where the respondents were shown example remarks from a random subset of clusters and got to evaluate both how well the extracted keywords matched the examples and to what degree the example remarks from the same cluster represented the same kind of error. We got a total of 22 responses consisting of 8 professional inspectors and 14 laymen. Our results show that the keyword extraction make sense in connection to the example remarks from the form and that the keywords show promise in describing the content of a cluster. Also, for a majority of the clusters a clear consensus can be seen between the respondents on what keywords they considered as relevant. However the average number of keywords that the respondents considered relevant for each remark (1.40) was deemed too low for us to be able to recommend the solution. Additionally the clustering quality follows the same pattern in showing promise but not quite giving satisfactory results in this study. For future work a larger study should be conducted where several combinations of clustering and keyword extraction methods could be evaluated more thoroughly to be able to draw more decisive conclusions. / Nuförtiden är det vanligt att företag samlar in och sitter på en mängd data kopplad till sin verksamhet. Denna datamängd är ofta för stor för att kunna analyseras för hand. Därför har det blivit allt vanligare att automatisera denna analys genom att köra maskininlärningsmetoder på datan. I detta projekt analyseras ett dataset bestående av fritext-poster innehållande anmärkningar på elinstallationer. Detta görs genom att först klustra datan med målet att varje kluster ska representera en viss typ av anmärkning från fältet för att sedan extrahera 10 st nyckelord från varje kluster. Vår undersökning går sedan ut på att undersöka till vilken grad dessa nyckelord kan sägas representera klustrens innehåll på ett sätt som skulle vara användbart för en applikation för slutanvändare. Den lösning som togs fram i projektet utvärderades genom en enkät där de svarande visades exempel på anmärkningar från ett antal slumpvist valda kluster och sedan fick ta ställning till hur väl nyckelorden passade in på exemplen och också till vilken grad exemplen från samma kluster representerade samma typ av anmärkning. Totalt fick vi in svar från 22 personer, nämligen 8 besiktningsingenjörer och 14 st lekmän. Resultaten visar att de extraherade nyckelorden hade en naturlig koppling till de respektive anmärkningarna från enkäten och att de har potential att förklara innehållet i klustren. Hos en majoritet av klustern kunde vi också se en tydlig samstämmighet bland de svarande i vilka specifika nyckelord som ansågs relevanta. Dock var det genomsnittliga antalet nyckelord som ansågs relevanta för ett anmärkningsexempel (1,40) för lågt för att vi ska kunna rekommendera den utvärderade lösningen. På ett liknande sätt visar våra resultat att klustringen av datan var lovande, men att den inte blev helt tillfredsställande. I ett fortsatt arbete borde en större undersökning göras där flera kombinationer av metoder för klustring och extrahering av nyckelord jämförs grundligare så att säkrare slutsatser kan dras.

Web-based learning should move away from static transmission of instruction to dynamic pages for effective interactive learning. Furthermore, automated assessment of learning should move beyond rigid quizzes or multiple-choice questions. This study describes the design, development, implementation, testing and evaluation of two prototypes of an electronic assessment tool to enhance the effectiveness of automated assessment. The tool was developed in the context of a distance-learning organisation and was built according to a development research model entailing a cyclic design-intervention-outcomes process. The first variant, E-Grader, was developed to test an algorithm for assigning marks to open-ended textual responses. The second variant, Web-Grader, was an interactive web-based extension of E-Grader. It provided immediate interactive support to students as they responded textually to content-based questions. This multi-disciplinary study incorporates principles and techniques from software engineering, formal computer science, database development and instructional design in the quest towards electronic assessment of web-based textual inputs. / Computing / M.Sc. (Information Systems)

Web-based learning should move away from static transmission of instruction to dynamic pages for effective interactive learning. Furthermore, automated assessment of learning should move beyond rigid quizzes or multiple-choice questions. This study describes the design, development, implementation, testing and evaluation of two prototypes of an electronic assessment tool to enhance the effectiveness of automated assessment. The tool was developed in the context of a distance-learning organisation and was built according to a development research model entailing a cyclic design-intervention-outcomes process. The first variant, E-Grader, was developed to test an algorithm for assigning marks to open-ended textual responses. The second variant, Web-Grader, was an interactive web-based extension of E-Grader. It provided immediate interactive support to students as they responded textually to content-based questions. This multi-disciplinary study incorporates principles and techniques from software engineering, formal computer science, database development and instructional design in the quest towards electronic assessment of web-based textual inputs. / Computing / M.Sc. (Information Systems)

資訊檢索(Information Retrieval)一直是資訊學界的重要研究領域，但長久以來並未能在其它的學門中發揮其重要性，然而藉由Internet的普及與網路資源的激增，以資訊檢索為基礎的網路搜尋技術，已逐漸成為Internet上相當受到重視的技術之一。雖然全文資訊檢索的技術已存在多時，然而電腦產業一直缺乏一廣泛標準的績效評估，來評估全文資訊檢索系統的產出及價格/效能比，1990年代對全文資訊檢索軟體的大量需求，極需要一個標準、一致的方法，來比較各系統間效能的差異。本研究即針對傳播中文全文資訊檢索資料庫系統為探討的主題，試圖以關連式資料庫績效評估的理論及方法，來建構其工作量模式以進行績效評估。 / Information Retrieval has been a important research domain in Information academic. But it does not play in the other domain for a long time. However the Internet search technology based on Information Retrieval has been an one of the important technologies in the Internet by the spreading of Internet and the growth of network resources. Although the Full-Text Information Retrieval or Free-Text Information Retrieval technology has existed for a long time, the computer industry was lack of a wide, standard benchmark to evaluate the throughput and price/performance of the Full-Text Information Retrieval System. The large demand for Full-Text Information Retrieval System or Free-Text Information Retrieval System software needs a standard method to compare the performance of each system in 90s. This research is focused on the Communication Free-Text Information Retrieval Database System in Chinese, and construct it’s workload model to evaluate by using the theory and method of Relational Database Benchmark.