Apprentissage automatique de la morphologie : le cas des structures racine-schème

RÉSUMÉ Cette thèse porte sur le développement de méthodes algorithmiques pour découvrir automatiquement la structure morphologique des mots d'un corpus. On considère en particulier le cas des langues s'approchant du type introflexionnel, comme l'arabe ou l'hébreu. La tradition linguistique décrit la morphologie de ces langues en termes d'unités discontinues : les racines consonantiques et les schèmes vocaliques. Ce genre de structure constitue un défi pour les systèmes actuels d'apprentissage automatique, qui opèrent généralement avec des unités continues. La stratégie adoptée ici consiste à traiter le problème comme une séquence de deux sous-problèmes. Le premier est d'ordre phonologique : il s'agit de diviser les symboles (phonèmes, lettres) du corpus en deux groupes correspondant autant que possible aux consonnes et voyelles phonétiques. Le second est de nature morphologique et repose sur les résultats du premier : il s'agit d'établir l'inventaire des racines et schèmes du corpus et de déterminer leurs règles de combinaison. On examine la portée et les limites d'une approche basée sur deux hypothèses : (i) la distinction entre consonnes et voyelles peut être inférée sur la base de leur tendance à alterner dans la chaîne parlée; (ii) les racines et les schèmes peuvent être identifiés respectivement aux séquences de consonnes et voyelles découvertes précédemment. L'algorithme proposé utilise une méthode purement distributionnelle pour partitionner les symboles du corpus. Puis il applique des principes analogiques pour identifier un ensemble de candidats sérieux au titre de racine ou de schème, et pour élargir progressivement cet ensemble. Cette extension est soumise à une procédure d'évaluation basée sur le principe de la longueur de description minimale, dans- l'esprit de LINGUISTICA (Goldsmith, 2001). L'algorithme est implémenté sous la forme d'un programme informatique nommé ARABICA, et évalué sur un corpus de noms arabes, du point de vue de sa capacité à décrire le système du pluriel. Cette étude montre que des structures linguistiques complexes peuvent être découvertes en ne faisant qu'un minimum d'hypothèses a priori sur les phénomènes considérés. Elle illustre la synergie possible entre des mécanismes d'apprentissage portant sur des niveaux de description linguistique distincts, et cherche à déterminer quand et pourquoi cette coopération échoue. Elle conclut que la tension entre l'universalité de la distinction consonnes-voyelles et la spécificité de la structuration racine-schème est cruciale pour expliquer les forces et les faiblesses d'une telle approche. ABSTRACT This dissertation is concerned with the development of algorithmic methods for the unsupervised learning of natural language morphology, using a symbolically transcribed wordlist. It focuses on the case of languages approaching the introflectional type, such as Arabic or Hebrew. The morphology of such languages is traditionally described in terms of discontinuous units: consonantal roots and vocalic patterns. Inferring this kind of structure is a challenging task for current unsupervised learning systems, which generally operate with continuous units. In this study, the problem of learning root-and-pattern morphology is divided into a phonological and a morphological subproblem. The phonological component of the analysis seeks to partition the symbols of a corpus (phonemes, letters) into two subsets that correspond well with the phonetic definition of consonants and vowels; building around this result, the morphological component attempts to establish the list of roots and patterns in the corpus, and to infer the rules that govern their combinations. We assess the extent to which this can be done on the basis of two hypotheses: (i) the distinction between consonants and vowels can be learned by observing their tendency to alternate in speech; (ii) roots and patterns can be identified as sequences of the previously discovered consonants and vowels respectively. The proposed algorithm uses a purely distributional method for partitioning symbols. Then it applies analogical principles to identify a preliminary set of reliable roots and patterns, and gradually enlarge it. This extension process is guided by an evaluation procedure based on the minimum description length principle, in line with the approach to morphological learning embodied in LINGUISTICA (Goldsmith, 2001). The algorithm is implemented as a computer program named ARABICA; it is evaluated with regard to its ability to account for the system of plural formation in a corpus of Arabic nouns. This thesis shows that complex linguistic structures can be discovered without recourse to a rich set of a priori hypotheses about the phenomena under consideration. It illustrates the possible synergy between learning mechanisms operating at distinct levels of linguistic description, and attempts to determine where and why such a cooperation fails. It concludes that the tension between the universality of the consonant-vowel distinction and the specificity of root-and-pattern structure is crucial for understanding the advantages and weaknesses of this approach.

Non-Destructive Evaluation of Iowa Pavements Phase 2: Development of a Fully Automated Software System for Rapid Analysis/Processing of the Falling Weight Deflectometer Data, 2009

The Office of Special Investigations at Iowa Department of Transportation (DOT) collects FWD data on regular basis to evaluate pavement structural conditions. The primary objective of this study was to develop a fully-automated software system for rapid processing of the FWD data along with a user manual. The software system automatically reads the FWD raw data collected by the JILS-20 type FWD machine that Iowa DOT owns, processes and analyzes the collected data with the rapid prediction algorithms developed during the phase I study. This system smoothly integrates the FWD data analysis algorithms and the computer program being used to collect the pavement deflection data. This system can be used to assess pavement condition, estimate remaining pavement life, and eventually help assess pavement rehabilitation strategies by the Iowa DOT pavement management team. This report describes the developed software in detail and can also be used as a user-manual for conducting simulation studies and detailed analyses. *********************** Large File ***********************

Pile Design and Tests for Integral Abutment Bridges, HR-273, 1987

Expansion joints increase both the initial cost and the maintenance cost of bridges. Integral abutment bridges provide an attractive design alternative because expansion joints are eliminated from the bridge itself. However, the piles in these bridges are subjected to horizontal movement as the bridge expands and contracts during temperature changes. The objective of this research was to develop a method of designing piles for these conditions. Separate field tests simulating a pile and a bridge girder were conducted for three loading cases: (1) vertical load only, (2) horizontal displacement of pile head only, and (3) combined horizontal displacement of pile head with subsequent vertical load. Both tests (1) and (3) reached the same ultimate vertical load, that is, the horizontal displacement had no effect on the vertical load capacity. Several model tests were conducted in sand with a scale factor of about 1:10. Experimental results from both the field and model tests were used to develop the vertical and horizontal load-displacement properties of the soil. These properties were input into the finite element computer program Integral Abutment Bridge Two-Dimensional (IAB2D), which was developed under a previous research contract. Experimental and analytical results compared well for the test cases. Two alternative design methods, both based upon the American Association of State Highway and Transportation Officials (AASHTO) Specification, were developed. Alternative One is quite conservative relative to IAB2D results and does not permit plastic redistribution of forces. Alternative Two is also conservative when compared to IAB2D, but plastic redistribution is permitted. To use Alternative Two, the pile cross section must have sufficient inelastic rotation capacity before local buckling occurs. A design example for a friction pile and an end-bearing pile illustrates both alternatives.

Feasibility of Using Automated Distress Data in the County Need Study, TR-418, 1998

The purpose of this project was to determine the feasibility of using pavement condition data collected for the Iowa Pavement Management Program (IPMP) as input to the Iowa Quadrennial Need Study. The need study, conducted by the Iowa Department of Transportation (Iowa DOT) every four years, currently uses manually collected highway infrastructure condition data (roughness, rutting, cracking, etc.). Because of the Iowa DOT's 10-year data collection cycles, condition data for a given highway segment may be up to 10 years old. In some cases, the need study process has resulted in wide fluctuations in funding allocated to individual Iowa counties from one study to the next. This volatility in funding levels makes it difficult for county engineers to plan and program road maintenance and improvements. One possible remedy is to input more current and less subjective infrastructure condition data. The IPMP was initially developed to satisfy the Intermodal Surface Transportation Efficiency Act (ISTEA) requirement that federal-aid-eligible highways be managed through a pavement management system. Currently all metropolitan planning organizations (MPOs) in Iowa and 15 of Iowa's 18 RPAs participate in the IPMP. The core of this program is a statewide data base of pavement condition and construction history information. The pavement data are collected by machine in two-year cycles. Using pilot areas, researchers examined the implications of using the automated data collected for the IPMP as input to the need study computer program, HWYNEEDS. The results show that using the IPMP automated data in HWYNEEDS is feasible and beneficial, resulting in less volatility in the level of total need between successive quadrennial need studies. In other words, the more current the data, the smaller the shift in total need.

Spatial behavior in groups: an agent-based approach

We present an agent-based model with the aim of studying how macro-level dynamics of spatial distances among interacting individuals in a closed space emerge from micro-level dyadic and local interactions. Our agents moved on a lattice (referred to as a room) using a model implemented in a computer program called P-Space in order to minimize their dissatisfaction, defined as a function of the discrepancy between the real distance and the ideal, or desired, distance between agents. Ideal distances evolved in accordance with the agent's personal and social space, which changed throughout the dynamics of the interactions among the agents. In the first set of simulations we studied the effects of the parameters of the function that generated ideal distances, and in a second set we explored how group macrolevel behavior depended on model parameters and other variables. We learned that certain parameter values yielded consistent patterns in the agents' personal and social spaces, which in turn led to avoidance and approaching behaviors in the agents. We also found that the spatial behavior of the group of agents as a whole was influenced by the values of the model parameters, as well as by other variables such as the number of agents. Our work demonstrates that the bottom-up approach is a useful way of explaining macro-level spatial behavior. The proposed model is also shown to be a powerful tool for simulating the spatial behavior of groups of interacting individuals.

Dynamic Pavement Deflection Measurements, Progress Report, HR-178, 1977

The Road Rater is a dynamic deflection measuring appa-ratus for flexible base pavements. The basic operating principle of the Road Rater is to impart a dynamic loading and measure the resultant movement of the pavement with velocity sensors. This data, when properly adjusted for temperature by use of a nomograph included in this report, can be used to determine pavement life expectancy and estimate overlay thickness required. Road Rater testing will be conducted in the spring, when pave-ments are in their weakest condition, until seasonal correction factors can be developed. The Road Rater does not have sufficient ram weight to effectively evaluate load carrying capacity of rigid pavements. All rigid pavements react similarly to Road Rater testing and generally deflect from 0.65 to 1.30 mils. Research will be continued to evaluate rigid pavements with the Road Rater, however. The Road Rater has proven to be a reliable, trouble free pavement evaluation machine. The deflection apparatus was originally front-mounted, but was rear-mounted during the winter of 1977-78. Since that time, van handling has greatly improved, and front suspension parts are no longer overstressed due to improper weight distribution. The Road Rater provides a fast, economical, nondestructive test method to evaluate flexible pavements. Road Rater test data can be used to predict pavement life, set priorities for asphaltic concrete resurfacing, and design asphaltic concrete overlays. Temperature and seasonal variations significantly affect Road Rater deflection readings and must be considered. A nomograph included in this report adjusts for temperature, but does not correct for seasonal effect. Road Rater testing will be conducted in the spring until seasonal correction factors can be developed. The Road Rater has not successfully evaluated rigid pavements, but research will continue in this area. 1. Recommendations for continuing Road Rater research, evaluation and application are as follows:A computer program should be established to reduce Road Rater raw data (Range and Sensor reading) to HR-178 Road Rater Dynamic Deflections For Determining Structural Rating Of Flexible Pavements mean deflection (mils) and/or structural rating. This computer printout would be similar to present friction testing printouts, and would greatly reduce Road Rater data reduction manpower needs and costs. 2. Seasonal variation study should continue to develop seasonal correction factors. Seasonal test roads will be studied concurrently with routine testing during 1979 to develop this relationship. All Road Rater testing will be conducted in the spring until the seasonal relationship is established. 3. An asphaltic concrete overlay design method should be established based on Road Rater de-flection readings. The AASHTO Interim Guide for Design of Pavement Structures 1972 will be used as a base document for this study. 4. AASHTO Structural numbers should be compared to Road Rater Structural Ratings during 1979 on asphaltic concrete overlay projects. This analysis will enable us to refine Road Rater evaluation of flexible pavements. Roads will be tested before resurfacing and several months

Iowa's CHLOE Profilometer Correlation Procedure, MLR-87-10, 1987

The Iowa DOT has been correlating its roadmeters to the CHLOE Profilometer since 1968. The same test method for the Present Serviceability Index (PSI) deduction from the pavement condition (crack and patch) survey has also been used since 1968. Resulting PSI measurements on the Interstate and Primary Highway Systems have had good continuity through the years due to these test procedures. A computer program called PSITREND has been developed to plot PSI versus year tested for every rural pavement section in the State of Iowa. PSITREND provides pavement performance trends which are very useful for prediction of rehabilitation needs and for evaluation of new designs or rehabilitation techniques. The PSITREND data base should be maintained through future years to expand on nineteen years of historical PSI test information already collected.

Factors Influencing Stability of Granular Base Course Mixes, Progress Report, HR-99, 1965

The problems of laboratory compaction procedures, the effect of gradation and mineralogy on shearing strength, and effect of stabilizing agents on shearing strength of granular base course mixes are discussed. For the materials tested, a suitable laboratory compaction procedure was developed which involves the use of a vibratory table to prepare triaxial test specimens. A computer program has been developed to facilitate the analysis of the test data of the effect of gradation and mineralogy on shearing strength of soils. The effects of the following materials have been selected for evaluation as stabilizing agents’ portland cement, sodium and calcium chloride, lime organic cationic waterproofer, and asphaltic materials.

Factors Influencing Stability of Granular Base Course Mixes, Progress Report, HR-99, 1964 (April)

The problems of laboratory compaction procedures, the effect of gradation and mineralogy on shearing strength, and effect of stabilizing agents on shearing strength of granular base course mixes are discussed. For the materials tested, a suitable laboratory compaction procedure was developed which involves the use of a vibratory table to prepare triaxial test specimens. A computer program has been developed to facilitate the analysis of the test data of the effect of gradation and mineralogy on shearing strength of soils. The effects of the following materials have been selected for evaluation as stabilizing agents’ portland cement, sodium and calcium chloride, lime organic cationic waterproofer, and asphaltic materials.

Factors Influencing Stability of Granular Base Course Mixes, Progress Report, HR-99, 1964 (November)

The problems of laboratory compaction procedures, the effect of gradation and mineralogy on shearing strength, and effect of stabilizing agents on shearing strength of granular base course mixes are discussed. For the materials tested, a suitable laboratory compaction procedure was developed which involves the use of a vibratory table to prepare triaxial test specimens. A computer program has been developed to facilitate the analysis of the test data of the effect of gradation and mineralogy on shearing strength of soils. The effects of the following materials have been selected for evaluation as stabilizing agents’ portland cement, sodium and calcium chloride, lime organic cationic waterproofer, and asphaltic materials.

R commands secondary production

R commands to calculate the secondary production estimates using the size-frequency method after Hynes and Coleman (1968), Benke (1979) and Huryn (1996).

Grau de satisfacció d'un col·lectiu de persones grans en relació a l'atenció rebuda dels professionals d'infermeria durant la seva hospitalització

Introducció: La preocupació per la qualitat assistencial, i concretament per la qualitat de les cures infermeres, la satisfacció percebuda pel pacient com element fonamental com a indicador de la qualitat dels serveis sanitaris i clau en la millora de la qualitat assistencial, i el fet que a l’estat espanyol hi ha pocs estudis de satisfacció específics de la població anciana hospitalitzada, ens va portar a plantejar un estudi amb l’objectiu de valorar el grau de satisfacció de les persones grans hospitalitzades en relació a l’atenció rebuda dels professionals d’infermeria. Metodologia: S’ha realitzat un estudi descriptiu i transversal en un col.lectiu de 42 persones hospitalitzades d’edat ≥ 60 anys. El grau de satisfacció es va determinar amb “La Monica Oberst Patient Satisfaction Scale 12” (LOPSS 12), versió reduïda en castellà. La personalització de les cures es va valorar amb un qüestionari que valora la relació infermera-usuari extret de l’Índex de Personalització de les Cures. L'anàlisi estadística es va dur a terme amb el programa informàtic SPSS® vs 21.0. Resultats: Es va evidenciar un elevat grau de satisfacció global amb les cures d’infermeria (8,8  1,5 / 10), però el nivell global de personalització d’aquestes cures no va ser molt satisfactori (3,4  2,1 / 7). Els homes van valorar millor a les infermeres respecte a que consideraven les seves opinions i preferències (p=0,03) i contestaven amb rapidesa les seves trucades (p=0,009). La personalització de les cures millorava conforme incrementava el grau de satisfacció global (r = 0,582; p = 0,000) i amb el temps d’hospitalització (r = 0,344; p = 0,026). Conclusions: El grau de satisfacció global respecte la satisfacció en l’atenció rebuda per part d’infermeria, va ser elevat però amb baix grau de personalització, mostrant major satisfacció els homes respecte les dones. Conèixer l’opinió dels usuaris representa una oportunitat pels professionals de reflexionar sobre la pràctica clínica i ser crítics per avançar en el desenvolupament de la professió, identificant aspectes susceptibles de millora 5 amb la finalitat d’augmentar la qualitat de les cures que es presten, així com la qualitat de vida de les persones.

Tietokoneohjelmien rajapintojen tekijänoikeussuoja

English summary: Copyright protection of computer program interfaces (s. 852-853)

Simulando una vía de comunicación: el tramo de la vía romana entre el Coll de Panissars y Girona

Mediante programa informático es posible simular las vías de comunicación existentes entre asentamientos humanos, siendo muchos los trabajos publicados y varias las formas de aproximación al problema. El objetivo del trabajo que nos ocupa ha consistido en la generación de un nuevo modelo a partir del cual fuera posible simular una vía de comunicación, con la salvedad de que en este caso ya existía una idea aproximada de cual sería su trayectoria a partir de trabajos de otros autores. Esta vía de comunicación es la vía Augusta, en el tramo que une el Coll de Panissars y la ciudad de Girona.

Stringless Portland Cement Concrete Paving, TR-490, 2004

This report describes results from a study evaluating the use of stringless paving using a combination of global positioning and laser technologies. CMI and Geologic Computer Systems developed this technology and successfully implemented it on construction earthmoving and grading projects. Concrete paving is a new area for considering this technology. Fred Carlson Co. agreed to test the stringless paving technology on two challenging concrete paving projects located in Washington County, Iowa. The evaluation was conducted on two paving projects in Washington County, Iowa, during the summer of 2003. The research team from Iowa State University monitored the guidance and elevation conformance to the original design. They employed a combination of physical depth checks, surface location and elevation surveys, concrete yield checks, and physical survey of the control stakes and string line elevations. A final check on profile of the pavement surface was accomplished by the use of the Iowa Department of Transportation Light Weight Surface Analyzer (LISA). Due to the speed of paving and the rapid changes in terrain, the laser technology was abandoned for this project. Total control of the guidance and elevation controls on the slip-form paver were moved from string line to global positioning systems (GPS). The evaluation was a success, and the results indicate that GPS control is feasible and approaching the desired goals of guidance and profile control with the use of three dimensional design models. Further enhancements are needed in the physical features of the slipform paver oil system controls and in the computer program for controlling elevation.