<p>The goal was to extract knowledge from data that is logged by the electronic system of</p><p>every Volvo truck. This allowed the evaluation of large populations of trucks without requiring additional measuring devices and facilities.</p><p>An evaluation cycle, similar to the knowledge discovery from databases model, was</p><p>developed and applied to extract knowledge from data. The focus was on extracting</p><p>information in the logged data that is related to the class labels of different populations,</p><p>but also supported knowledge extraction inherent from the given classes. The methods</p><p>used come from the field of unsupervised learning, a sub-field of machine learning and</p><p>include the methods self-organizing maps, multi-dimensional scaling and fuzzy c-means</p><p>clustering.</p><p>The developed evaluation cycle was exemplied by the evaluation of three data-sets.</p><p>Two data-sets were arranged from populations of trucks differing by their operating</p><p>environment regarding road condition or gross combination weight. The results showed</p><p>that there is relevant information in the logged data that describes these differences</p><p>in the operating environment. A third data-set consisted of populations with different</p><p>engine configurations, causing the two groups of trucks being unequally powerful.</p><p>Using the knowledge extracted in this task, engines that were sold in one of the two</p><p>configurations and were modified later, could be detected.</p><p>Information in the logged data that describes the vehicle's operating environment,</p><p>allows to detect trucks that are operated differently of their intended use. Initial experiments</p><p>to find such vehicles were conducted and recommendations for an automated</p><p>application were given.</p>

The goal was to extract knowledge from data that is logged by the electronic system of every Volvo truck. This allowed the evaluation of large populations of trucks without requiring additional measuring devices and facilities. An evaluation cycle, similar to the knowledge discovery from databases model, was developed and applied to extract knowledge from data. The focus was on extracting information in the logged data that is related to the class labels of different populations, but also supported knowledge extraction inherent from the given classes. The methods used come from the field of unsupervised learning, a sub-field of machine learning and include the methods self-organizing maps, multi-dimensional scaling and fuzzy c-means clustering. The developed evaluation cycle was exemplied by the evaluation of three data-sets. Two data-sets were arranged from populations of trucks differing by their operating environment regarding road condition or gross combination weight. The results showed that there is relevant information in the logged data that describes these differences in the operating environment. A third data-set consisted of populations with different engine configurations, causing the two groups of trucks being unequally powerful. Using the knowledge extracted in this task, engines that were sold in one of the two configurations and were modified later, could be detected. Information in the logged data that describes the vehicle's operating environment, allows to detect trucks that are operated differently of their intended use. Initial experiments to find such vehicles were conducted and recommendations for an automated application were given.

Data Structures and Algorithms (DSA) courses are critical for any computer science curriculum. DSA courses emphasize concepts related to procedural dynamics and Algorithm Analysis (AA). These concepts are hard for students to grasp when conveyed using traditional textbook material relying on text and static images. Algorithm Visualizations (AVs) emerged as a technique for conveying DSA concepts using interactive visual representations. Historically, AVs have dealt with portraying algorithm dynamics, and the AV developer community has decades of successful experience with this. But there exist few visualizations to present algorithm analysis concepts. This content is typically still conveyed using text and static images. We have devised an approach that we term Algorithm Analysis Visualizations (AAVs), capable of conveying AA concepts visually. In AAVs, analysis is presented as a series of slides where each statement of the explanation is connected to visuals that support the sentence. We developed a pool of AAVs targeting the basic concepts of AA. We also developed AAVs for basic sorting algorithms, providing a concrete depiction about how the running time analysis of these algorithms can be calculated. To evaluate AAVs, we conducted a quasi-experiment across two offerings of CS3114 at Virginia Tech. By analyzing OpenDSA student interaction logs, we found that intervention group students spent significantly more time viewing the material as compared to control group students who used traditional textual content. Intervention group students gave positive feedback regarding the usefulness of AAVs to help them understand the AA concepts presented in the course. In addition, intervention group students demonstrated better performance than control group students on the AA part of the final exam. The final exam taken by both the control and intervention groups was based on a pilot version of the Algorithm Analysis Concept Inventory (AACI) that was developed to target fundamental AA concepts and probe students' misconceptions about these concepts. The pilot AACI was developed using a Delphi process involving a group of DSA instructors, and was shown to be a valid and reliable instrument to gauge students' understanding of the basic AA topics. / Ph. D.

Les forêts tropicales constituent le principal réservoir de biodiversité et de carbone (C). Cependant, la plupart des forêts tropicales, en particulier les forêts de Bornéo en Asie du Sud-Est, subissent une pression intense et sont menacées par des activités anthropiques telles que l'exploitation forestière, l'industrie minière l’agriculture et la conversion en plantations industrielles. En 2010, la superficie des forêts de production de Bornéo était de 26,8 millions d’ha (environ 36% de la superficie totale de l’île, dont 18 millions ha (environ 24%) déjà exploités. Par conséquent, les forêts de production occupent donc une place importante à Bornéo et jouent un rôle essentiel dans la compensation des biens fournis et la maintenance des services écosystémiques, tels que la conservation du C et de la biodiversité.L’exploitation sélective réduit la biomasse aérienne et souterraine par l’élimination de quelques grands arbres, et augmente les stocks de bois mort par des dommages collatéraux. En créant des trouées dans la canopée, le microclimat dans les sous-étages et au sol change localement et accélèrent la décomposition de la litière et de la matière organique. L'importance des dégâts, de l'ouverture de la canopée et de la rapidité du rétablissement du C s'est avéré principalement liée à l'intensité de l'exploitation forestière. Cependant, les évaluations empiriques de l'effet à long terme de l'intensité de l'exploitation forestière sur l'équilibre du C dans les forêts de production restent rares.La présente thèse se concentre principalement sur l'évaluation de l'effet à long terme de l'intensité de l'exploitation forestière sur la séquestration de carbone dans une forêt à Diptérocarpées de Nord Bornéo (District de Malinau, Kalimantan Nord) exploitée en 1999/2000. Cinq principaux réservoirs de C, à savoir le C aérien dans les arbres vivants (AGC), le C souterrain dans les arbres vivants (BGC), le bois mort, la litière et le C organique du sol (SOC) ont été estimés le long d’un gradient d'intensité d'exploitation (0-57% de la biomasse perdue).Nos résultats ont montré que les stocks totaux de C, 16 ans après l'exploitation, variaient de 218 à 554 Mg C ha-1 avec une moyenne de 314 Mg C ha-1. Une différence de 95 Mg C ha-1 a été observée entre une faible intensité d'exploitation forestière (<2,1% de la biomasse initiale perdue) et une intensité d'exploitation élevée (>19%). La plus grande partie du C (environ 77%) était présente dans les arbres vivants, suivie par les stocks du sol (15%), les stocks de bois mort (6%) et une fraction mineure des stocks de litière (1%). L'empreinte de l'intensité de l'exploitation forestière était encore détectable 16 ans après l'exploitation et a été le principal facteur expliquant la réduction des AGC>20, BGC>20, du bois mort et des stocks de C et une augmentation du bois mort. L'intensité de l'exploitation expliquait à elle seule 61%, 63%, 38% et 48% des variations des AGC>20, BGC>20, du bois mort et des stocks de C totaux, respectivement. L'intensité de l'abattage a également réduit considérablement les stocks de SOC dans la couche supérieure de 30 cm. Pour l'ensemble des stocks de SOC (0-100 cm), l'influence de l'intensité de l'exploitation était encore perceptible, en conjonction avec d'autres variables.Nos résultats quantifient l'effet à long terme de l'exploitation forestière sur les stocks de C forestier, en particulier sur les AGC et les bois morts. L'intensité élevée de l'exploitation forestière (réduction de 50% de la biomasse initiale) a réduit les stocks totaux de C de 27%. La récupération de l'AGC était plus faible dans les parcelles d'intensité d'exploitation forestière élevée, ce qui suggère une résilience plus faible de la forêt à l'exploitation forestière. Par conséquent, une intensité d'exploitation forestière inférieure à 20%, devrait être envisagé afin de limiter l'effet à long terme sur les AGC et le bois mort. / Tropical forests are a major reservoir of biodiversity and carbon (C), playing a pivotal role in global ecosystem function and climate regulation. However, most of the tropical forests, especially Bornean forests in Southeast Asia, are under intense pressure and threatened by anthropogenic activities such as logging, mining industry, agriculture and conversion to industrial plantation. In 2010, the area of production forests in Borneo was 26.8 million ha (approx. 36% of the total land area of Borneo) including 18 million ha (approx. 24%) of logged forests. Production forests are thus emerging as a dominant land-use, playing a crucial role in trading-off provision of goods and maintenance of ecosystem services, such as C and biodiversity retention.Selective logging is known to reduce both above- and below-ground biomass through the removal of a few large trees, while increasing deadwood stocks through collateral damages. By creating large gaps in the canopy, microclimates in the understory and on the forest floor change locally speeding up the decomposition of litter and organic matter. The extent of incidental damages, canopy openness, as well as the speed of C recovery, was shown to be primarily related to logging intensity. However, empirical evaluations of the long-term effect of logging intensity on C balance in production forests remain rare.The present thesis aims to assess the long-term effect of logging intensity on C sequestration in a north Bornean Dipterocarp forests (Malinau District, North Kalimantan) logged in 1999/2000. Five main C pools, namely above-ground (AGC) and below-ground (BGC) carbon in living trees, deadwood, litter, and soil organic carbon (SOC) were estimated along a logging intensity gradient (ranging from 0 to 57% of initial biomass removed).Our result showed that total C stocks 16 years after logging, ranged from 218-554 Mg C ha-1 with an average of 314 Mg C ha-1. A difference of 95 Mg C ha-1 was found between low logging intensity (<2.1% of initial biomass lost) and high logging intensity (>19%). Most C (approx. 77%) was found in living trees, followed by soil (15%), deadwood (6%), and a minor fraction in litter (1%). The imprint of logging intensity was still detectable 16 years after logging, and logging intensity thus was the main driver explaining the reduction of AGC>20, BGC>20, deadwood, and total C stocks and an increase in deadwood. Solely, logging intensity explained 61%, 63%, 38%, and 48% of variations of AGC>20, BGC>20, deadwood, and total C stocks, respectively. Logging intensity also significantly reduced SOC stocks in the upper 30 cm layer. For total SOC stocks (0-100 cm), the negative influence of logging intensity was still perceptible, being significant in conjunction with other variables.Our results quantify the long-term effect of logging on forest C stocks, especially on AGC and deadwood. High logging intensity (50% reduction of initial biomass) reduced total C stocks by 27%. AGC recovery was lower in high logging intensity plots, suggesting lowered forest resilience to logging. Our study showed that maintaining logging intensity, below 20% of the initial biomass, limit the long-term effect of logging on AGC and deadwood stocks.

This study reports forest structure, regeneration and the soil properties from unlogged and logged forest in the Bulungan Research Forest, Malinau District, East Kalimantan, Indonesia. Four sites were compared by using four 1-ha replicate plots in each of primary forest (PF), 5, 10 and 30-yr old logged forest (LF-5, LF-10, LF- 30). The tree species composition differ among forest types, as it was shown that the mean value of similarity indices for all pairs were 0.215 (for the Jaccard index) and 0.353 (for the Sorensen index). The low values for similarities among forest types were most probably caused by low numbers of species shared between each forest type. Both correlation values, r = 0.023 for Jaccard index and r = 0.031 for Sorensen index, showed no strong correlation between the similarity index (C) and the distance between forest types. This supports the use of a chronosequence approach. A total of 914 tree species with ³ 10 cm dbh were recorded from 223 genera and 65 families. There were no significant differences in mean species numbers (166 – 180/ha) among treatments. Mean density of species was lower in LF-5 and LF-10 (501/ha) than in PF or LF-30 (605/ha and 577/ha); similarly to mean basal area (LF-5, 28.5 m2/ha; LF-10, 32.6 m2/ha) vs. PF (45.8 m2/ha) and LF-30 (46.9 m2/ha). Dead wood on the forest floor was significantly higher in LF-10 (75 m3/ha) than in the other treatments. Seedlings (< 2 cm dbh) of 1,022 species were recorded from 408 genera and 111 families. The mean number of tree seedling species ranged between 170-206; the mean density of seedlings was about two-fold lower in LF-10 (2790/ha) than in the other treatments. Saplings (>2 – 9.9 cm dbh) of 802 species belonged to 241 genera and 65 families. There was a high variability in species richness across treatments (89 – 191/ha), but not in stem numbers. The Dipterocarpaceae family was dominant in all treatments, followed by the Euphorbiaceae. The soils were acidic, low in nutrients and had low to very low fertility. Both primary and logged forest areas are marginal or not suitable for sustained production of plantation crops. Logging caused soil compaction in LF-30. Although in terms of number of species and trees, amount of BA, number of saplings and seedlings LF-30 appeared to have satisfied prescriptions for a second harvest, ecologically the forest is far from mature. The Indonesian Selective Cutting and Replanting (TPTI) system may need to be revised to a 35 – 45 year cycle to ensure long-term forest productivity in terms of not only timber but other goods and ecosystem services, the value of which are never quantified in monetary terms, but can be higher than the timber revenue.

The large potential of big data and how it has brought value into various industries have been established in research. Since big data has such large potential if handled and analyzed in the right way, revealing information to support decision making in an organization, this thesis is conducted as a case study at an automotive manufacturer with access to large amounts of customer usage data of their vehicles. The reason for performing an analysis of this kind of data is based on the cornerstones of Total Quality Management with the end objective of increasing customer satisfaction of the concerned products or services. The case study includes a data analysis exploring how and if patterns about what affects fuel consumption can be revealed from aggregated customer usage data of trucks linked to truck applications. Based on the case study, conclusions are drawn about how a company can use this type of analysis as well as how to handle the data in order to turn it into business value. The data analysis reveals properties describing truck usage using Factor Analysis and Principal Component Analysis. Especially one property is concluded to be important as it appears in the result of both techniques. Based on these properties the trucks are clustered using k-means and Hierarchical Clustering which shows groups of trucks where the importance of the properties varies. Due to the homogeneity and complexity of the chosen data, the clusters of trucks cannot be linked to truck applications. This would require data that is more easily interpretable. Finally, the importance for fuel consumption in the clusters is explored using model estimation. A comparison of Principal Component Regression (PCR) and the two regularization techniques Lasso and Elastic Net is made. PCR results in poor models difficult to evaluate. The two regularization techniques however outperform PCR, both giving a higher and very similar explained variance. The three techniques do not show obvious similarities in the models and no conclusions can therefore be drawn concerning what is important for fuel consumption. During the data analysis many problems with the data are discovered, which are linked to managerial and technical issues of big data. This leads to for example that some of the parameters interesting for the analysis cannot be used and this is likely to have an impact on the inability to get unanimous results in the model estimations. It is also concluded that the data was not originally intended for this type of analysis of large populations, but rather for testing and engineering purposes. Nevertheless, this type of data still contains valuable information and can be used if managed in the right way. From the case study it can be concluded that in order to use the data for more advanced analysis a big-data plan is needed at a strategic level in the organization. The plan summarizes the suggested solution for the managerial issues of the big data for the organization. This plan describes how to handle the data, how the analytic models revealing the information should be designed and the tools and organizational capabilities needed to support the people using the information.