El presente proyecto de investigación es un análisis desde la perspectiva de negocios y estadístico de la empresa Agrokasa. El objetivo principal es encontrar nuevos mercados alternativos en crecimiento que le permitan obtener una mejor rentabilidad por el precio de kilo exportado de palta. Para alcanzar lo mencionado, se llevó a cabo un análisis empresarial que nos permita comprender el contexto y rubro de la empresa. Una vez alcanzado ese objetivo, se aplicó la metodología de la ciencia de datos para encontrar países de destino que son atractivos para Agrokasa. En cuanto al conjunto de datos, se obtuvo de diferentes fuentes públicas y privadas como Veritrade, Trade Map y Adex Data Trade. En consecuencia, se logró identificar 03 mercados alternativos y potenciales, tales como, Rusia, China y Corea Del Sur. En el análisis se utilizaron diferentes herramientas tecnológicas para la compilación, depuración, procesamiento y visualización de los datos, tales como Excel, Power Bi y Python. Con lo cual se demostró la importancia de ver todas las variables en una visualización que nos permite entender el comportamiento de los datos y nos sirve como fundamento para la toma de decisiones. En cuanto a los nuevos mercados, China presento el mayor valor total FOB exportado en el periodo analizado, 2018 -2020. Pese a presentar una tendencia negativa en la Regresión Lineal. Sin embargo, el precio promedio por kilo de palta aun es conveniente. Por otro lado, Rusia fue el mercado con mayores perspectivas de crecimiento y Corea Del Sur con un mejor precio por KG. Finalmente, para todos los mercados se utilizó una técnica de ciencia de datos con aprendizaje supervisado con un enfoque predictivo para pronosticar las importaciones de cada uno de ellos a fin de establecer estrategias comerciales para penetrar en ellos. / This paper is an analysis from a business and statistical perspective of the Agrokasa company in order to find new potential markets that allow it to grow in the volume of its avocado exports and in profitability per Kg exported. To achieve the aforementioned, a previous analysis from a business approach has been used, to understand the context and business area. Once this is understood, the methodology of data science has been applied to find destination countries that are attractive to Agrokasa. The data set was obtained from different public and private sources such as Veritrade and Trademap, with which it was possible to identify 03 potential markets that were China, Russia and South Korea. In the analysis, different technological tools were used to compile, debug, process and visualize the data, such as Excel, Power Bi and Python. With which it was demonstrated the importance of seeing all the variables in a visualization that allows us to understand the behavior of the data and serves as a basis for decision-making. China was the market with the highest total FOB value exported in the analyzed period, which was from 2018 -2020, however, with a negative trend, but with a convenient average price. On the other hand, Russia was the market with the best growth prospects and South Korea with a better price per KG. ​Finally, for all markets, a data science technique with supervised learning with a predictive approach was used to forecast the imports of each of them in order to establish commercial strategies to penetrate them. / Trabajo de investigación

La presente investigación propone un modelo para la automatización de predicción de la deserción de estudiantes universitarios. Esta investigación surge de una problemática existente en el sector educativo peruano: la deserción estudiantil universitaria; es decir, aquellos estudiantes universitarios que abandonan sus estudios de forma parcial o definitiva. La investigación tiene por finalidad brindar una solución que contribuya a reducir la tasa de deserción universitaria, aplicando tecnologías de análisis predictivo y minería de datos, que detecte anticipadamente a estudiantes con posibilidades de abandonar sus estudios, brindando así a las instituciones educativas mayor visibilidad y oportunidades de acción ante esta problemática. Se diseñó un modelo de análisis predictivo, en base al análisis y definición de 15 variables de predicción, 3 fases y la aplicación de algoritmos de predicción, basados en la disciplina del Educational Data Minig (EDM) y soportada por la plataforma IBM SPSS Modeler. Para validar, se evaluó la aplicación de 4 algoritmos de predicción: árboles de decisión, redes bayesianas, regresión lineal y redes neuronales; en un estudio en una institución universitaria de Lima. Los resultados indican que las redes bayesianas se comportan mejor que otros algoritmos, comparados bajo las métricas de precisión, exactitud, especificidad y tasa de error. Particularmente, la precisión de las redes bayesianas alcanza un 67.10% mientras que para los árboles de decisión (el segundo mejor algoritmo) es de un 61,92% en la muestra de entrenamiento para la iteración con razón de 8:2. Además, las variables “persona deportista” (0,29%), “vivienda propia” (0,20%) y “calificaciones de preparatoria” (0,15%) son las que más contribuyen al modelo de predicción. / This research proposes a model for the automation of prediction of university student dropout. This research arises from an existing problem in the Peruvian educational sector: university student dropout; that is, those university students who partially or permanently abandon their studies. The purpose of the research is to provide a solution that contributes to reducing the university dropout rate, applying predictive analysis technologies and data mining, which detects in advance students with the possibility of dropping out of their studies, thus providing educational institutions with greater visibility and opportunities. of action before this problem. A predictive analysis model was designed, based on the analysis and definition of 15 prediction variables, 3 phases and the application of prediction algorithms, based on the Educational Data Mining (EDM) discipline and supported by the IBM SPSS Modeler platform. To validate, the application of 4 prediction algorithms was evaluated: decision trees, Bayesian networks, linear regression, and neural networks; in a study at a university institution in Lima. The results indicate that Bayesian networks perform better than other algorithms, compared under the metrics of precision, accuracy, specificity, and error rate. Particularly, the precision of Bayesian networks reaches 67.10% while for decision trees (the second-best algorithm) it is 61.92% in the training sample for the iteration with a ratio of 8: 2. In addition, the variables "sports person" (0.29%), "own home" (0.20%) and "high school grades" (0.15%) are the ones that contribute the most to the prediction model. / Tesis

No description available.

No description available.

Projektet genomfördes i samarbete med Siemens Healthineers i syfte att utreda möjligheter till att prognostisera vårdflöden. Det genom att undersöka hur big data tillsammans med maskininlärning kan utnyttjas för prediktiv analys. Projektet utgjordes av två fallstudier med mål att, baserat på data från tidigare MRT-undersökningar, förutspå undersökningstider för kommande undersökningar respektive identifiera patienter som riskerar att missa inbokad undersökning. Fallstudierna utfördes med hjälp av programmeringsspråket R och tre olika inbyggda funktioner för maskininlärning användes för att ta fram prediktiva modeller för respektive fallstudie. Resultaten från fallstudierna gav en indikation på att det med en större datamängd av bättre kvalitet skulle vara möjligt att förutspå undersökningstider och vilka patienter som riskerar att missa sin inbokade undersökning. Det talar för att den här typen av prediktiva analyser kan användas för att prognostisera vårdflöden, något som skulle kunna bidra till ökad effektivitet och kortare väntetider i vården. / This project was performed in cooperation with Siemens Healthineers. The project aimed to investigate possibilities to forecast healthcare processes by investigating how big data and machine learning can be used for predictive analytics. The project consisted of two separate case studies. Based on data from previous MRI examinations the aim was to investigate if it is possible to predict duration of MRI examinations and identify potential no show patients. The case studies were performed with the programming language R and three machine learning methods were used to develop predictive models for each case study. The results from the case studies indicate that with a greater amount of data of better quality it would be possible to predict duration of MRI examinations and potential no show patients. The conclusion is that these types of predictive models can be used to forecast healthcare processes. This could contribute to increased effectivity and reduced waiting time in healthcare.

Sport clubs are always searching for innovative ways to improve performance and obtain a competitive edge. Sports analytics today is focused primarily on evaluating metrics thought to be directly tied to performance. Injuries indirectly decrease performance and cost substantially in terms of wasted salaries. Existing sports injury research mainly focuses on correlating one specific feature at a time to the risk of injury. This paper provides a multidimensional approach to non-contact injury prediction in Swedish professional ice hockey by applying machine learning on historical data. Several features are correlated simultaneously to injury probability. The project’s aim is to create an injury predicting algorithm which ranks the different features based on how they affect the risk of injury. The paper also discusses the business potential and strategy of a start-up aiming to provide a solution for predicting injury risk through statistical analysis. / Idrottsklubbar letar ständigt efter innovativa sätt att förbättra prestation och erhålla konkurrensfördelar. Idag fokuserar data- analys inom idrott främst på att utvärdera mätvärden som tros vara direkt korrelerade med prestation. Skador sänker indirekt prestationen och kostar markant i bortslösade spelarlöner. Tidigare studier på skador inom idrotten fokuserar huvudsakligen på att korrelera ett mätvärde till en skada i taget. Den här rapporten ger ett multidimensionellt angreppssätt till att förutse skador inom svensk elitishockey genom att applicera maskininlärning på historisk data. Flera attribut korreleras samtidigt för att få fram en skadesannolikhet. Målet med den här rapporten är att skapa en algoritm för att förutse skador och även ranka olika attribut baserat på hur de påverkar skaderisken. I rapporten diskuteras även affärsmöjligheterna för en sådan lösning och hur en potentiell start-up ska positionera sig på marknaden.

The increasing demand for healthcare services, coupled with the challenges of managing budgets and navigating complex regulations, has underscored the need for sustainable and efficient healthcare delivery. In response to this pressing issue, this thesis aims to optimize hospital efficiency using Artificial Intelligence (AI) techniques. The focus extends beyond improving surgical intraoperative time to encompass preoperative and postoperative periods as well. The research presents a novel Prescriptive Analytics System (PAS) designed to enhance the Surgical Success Rate (SSR) in surgeries and specifically in high volume arthroplasty. The SSR is a critical metric that reflects the successful completion of 4-surgeries during an 8-hour timeframe. By leveraging AI, the developed PAS has the potential to significantly improve the SSR from its current rate of 39% at The Ottawa Hospital to a remarkable 100%. The research is structured around five peer-reviewed journal papers, each addressing a specific aspect of the optimization of surgical efficiency. The first paper employs descriptive analytics to examine the factors influencing delays and overtime pay during surgeries. By identifying and analyzing these factors, insights are gained into the underlying causes of surgery inefficiencies. The second paper proposes three frameworks aimed at improving Operating Room (OR) throughput. These frameworks provide structured guidelines and strategies to enhance the overall efficiency of surgeries, encompassing preoperative, intraoperative, and postoperative stages. By streamlining the workflow and minimizing bottlenecks, the proposed frameworks have the potential to significantly optimize surgical operations. The third paper outlines a set of actions required to transform a selected predictive system into a prescriptive one. By integrating AI algorithms with decision support mechanisms, the system can offer actionable recommendations to surgeons during surgeries. This transformative step holds tremendous potential in enhancing surgical outcomes while reducing time. The fourth paper introduces a benchmarking and monitoring system for the selected framework that predicts SSR. Leveraging historical data, this system utilizes supervised machine learning algorithms to forecast the likelihood of successful outcomes based on various surgical team and procedural parameters. By providing real-time monitoring and predictive insights, surgeons can proactively address potential risks and improve decision-making during surgeries. Lastly, an application paper demonstrates the practical implementation of the prescriptive analytics system. The case study highlights how the system optimizes the allocation of resources and enables the scheduling of additional surgeries on days with a high predicted SSR. By leveraging the system's capabilities, hospitals can maximize their surgical capacity and improve overall patient care.

<p>  </p> <p>During the construction of a bridge, more specifically a concrete bridge deck, there are sometimes defects in materials or workmanship, resulting in what is called a construction defect. These defects can have a large impact on the lifecycle performance of the bridge deck, potentially leading to more preventative and reactive maintenance actions over time and thus a larger monetary investment by the bridge owner. Bridge asset managers utilize prediction software to inform their annual budgetary needs, however this prediction software traditionally relies only on historical condition rating data for its predictions. When attempting to understand how deterioration of a bridge deck changes with the influence of construction defects, utilizing the current prediction software is not appropriate as there is not enough historical data available to ensure accuracy of the prediction. There are numerical modeling approaches available that capture the internal physical and chemical deterioration processes, and these models can account for the change in deterioration when construction defects are present. There are also numerical models available that capture the effect of external factors that may be affecting the deterioration patterns of the bridge deck, in parallel to the internal processes. The goal of this study is to combine a mechanistic model capturing the internal physical and chemical processes associated with deterioration of a concrete bridge deck, with a model that is built strictly from historical condition rating data, in order to predict the changes in condition rating prediction of a bridge deck for a standard construction case versus a substandard construction case. Being able to measure the change in prediction of deterioration when construction defects are present then allows for quantifying the additional cost that would be required to maintain the defective bridge deck which is also presented. </p>

In the data-driven landscape of modern business, predictive analytics plays a pivotal role inanticipating and mitigating customer churn—a critical challenge for organizations. However, thetraditional complexities of machine learning hinder accessibility for decision-makers. EnterMachine Learning as a Service (MLaaS), offering a gateway to predictive modeling without theneed for extensive coding or infrastructure.This thesis presents a comprehensive evaluation of cloud-based and cloud-agonostic AutoML(Automated Machine Learning) platforms for customer churn prediction. The study focuses onfour prominent platforms: Azure ML, AWS SageMaker, GCP Vertex AI, and Databricks. Theevaluation encompasses various performance metrics including accuracy, AUC-ROC, precision,recall to assess the predictive capabilities of each platform. Furthermore, the ease of use andlearning curve for model development are compared, considering factors such as data preparation,training steps, and coding requirements. Additionally, model training times are analyzed toidentify platform efficiencies. Preliminary results indicate that AWS SageMaker exhibits thehighest accuracy, suggesting strong predictive capabilities. GCP Vertex AI excels in AUC,indicating robust discriminatory power. Azure ML demonstrates a balanced performance,achieving notable accuracy and AUC scores. Databricks being platform independent is a winnerand has also shown good metrics. Its capability to generate notebook is an added advantage whichcan be modified by experts to fine tune the results more. This research provides valuable insightsfor organizations seeking to implement different AutoML solutions for customer churnprediction.

This thesis examines predictive modelling from the legal perspective. Predictive modelling is a technology based on applied statistics, mathematics, machine learning and artificial intelligence that uses algorithms to analyse big data collections, and identify patterns that are invisible to human beings. The accumulated knowledge is incorporated into computer models, which are then used to identify and predict human activity in new circumstances, allowing for the manipulation of human behaviour. Predictive models use big data to represent people. Big data is a term used to describe the large amounts of data produced in the digital environment. It is growing rapidly due mainly to the fact that individuals are spending an increasing portion of their lives within the on-line environment, spurred by the internet and social media. As individuals make use of the on-line environment, they part with information about themselves. This information may concern their actions but may also reveal their personality traits. Predictive modelling is a powerful tool, which private companies are increasingly using to identify business risks and opportunities. They are incorporated into on-line commercial decision-making systems, determining, among other things, the music people listen to, the news feeds they receive, the content people see and whether they will be granted credit. This results in a number of potential harms to the individual, especially in relation to personal autonomy. This thesis examines the harms resulting from predictive modelling, some of which are recognized by traditional law. Using the European legal context as a point of departure, this study ascertains to what extent legal regimes address the use of predictive models and the threats to personal autonomy. In particular, it analyses Article 8 of the European Convention on Human Rights (ECHR) and the forthcoming General Data Protection Regulation (GDPR) adopted by the European Union (EU). Considering the shortcomings of traditional legal instruments, a strategy entitled ‘empowerment’ is suggested. It comprises components of a legal and technical nature, aimed at levelling the playing field between companies and individuals in the commercial setting. Is there a way to strengthen humanity as predictive modelling continues to develop?