The goal of the master thesis project was to develop an application for tracking of actual events in the surrounding area of the users. This application should allow the users to view events, create new events and add comments to existing ones. Beyond the implementation of developed application, this project deals with an analysis of the presented problem. The analysis includes a comparison with existing solutions and search for available technologies and frameworks applicable for implementation. Another part inside this work is description of the theory in behind of data classification that is internally used for event and comment analysis. This work also includes a design of appliction including design of user interface, software architecture, database, communication protocol and data classifiers. The main part of this project, the implementation, is described aftewards. At the end of this work, there is a summary of the whole process and also there are given some ideas about enhancing the application in the future.

This work deals with methods of web page classification. It explains the concept of classification and different features of web pages used for their classification. Further it analyses representation of a page and in detail describes classification method that deals with hierarchical category model and is able to dynamically create new categories. In the second half it shows implementation of chosen method and describes the results.

This master's thesis presents methods for mining important pieces of information from text. It analyses the problem of terms extraction from large document collection and describes the implementation using C# language and Microsoft SQL Server. The system uses stemming and a number of statistical methods for term extraction. This project also compares used methods and suggests the process of the dictionary derivation.

This work addresses the need for an alternative to keyword-based search for sifting through large PDF medical journal article document collections for literature review purposes. Despite users' best efforts to form precise and accurate queries, it is often difficult to guess the right keywords to find all the related articles while finding a minimum number of unrelated ones. Failure during literature review to find relevant, related research results in wasted research time and effort in addition to missing significant work in the related area which could affect the quality of the research work being conducted. The purpose of this work is to explore the benefits of a retrieval system for professional journal articles in PDF format that supports hybrid queries composed of both text and images. PDF medical journal articles contain formatting and layout information that imply the structure and organization of the document. They also contain figures and tables rich with content and meaning. Stripping a PDF into “full-text” for indexing purposes disregards these important features. Specifically, this work investigated the following: (1) what effect the incorporation of a document's embedded figures into the query (in addition to its text) has on retrieval performance (precision) compared to plain keyword-based search; (2) how current text-based document-query similarity methods can be enhanced by using formatting and font-size information as a structure and organization model for a PDF document; (3) whether to use the standard Euclidean distance function or the matrix distance function for content-based image retrieval; (4) how to convert a PDF into a structured, formatted, reflowable XML representation given a pure-layout PDF document; (5) what document views (such as a term frequency cloud, a document outline, or a document's figures) would help users wade through search results to quickly select those that are worth a closer look. While the results of the experiments were unexpectedly worse than their baselines of comparison (see the conclusion for a summary), the experimental methods are very valuable in showing others what directions have already been pursued and why they did not work and what remaining problems need to be solved in order to achieve the goal of improving literature review through use of a hybrid text and image retrieval system.

Background: Movies have become an important marketing and advertising tool that can influence consumer behaviour and trends. Reading film reviews is an im- important part of watching a movie, as it can help viewers gain a general under- standing of the film. And also, provide filmmakers with feedback on how their work is being received. Sentiment analysis is a method of determining whether a review has positive or negative sentiment, and this study investigates a machine learning method for classifying sentiment from film reviews. Objectives: This thesis aims to perform comparative sentiment analysis on textual IMDb movie reviews using lexicon-based and BERT neural network models. Later different performance evaluation metrics are used to identify the most effective learning model. Methods: This thesis employs a quantitative research technique, with data analysed using traditional machine learning. The labelled data set comes from an online website called Kaggle (https://www.kaggle.com/datasets), which contains movie review information. Algorithms like the lexicon-based approach and the BERT neural networks are trained using the chosen IMDb movie reviews data set. To discover which model performs the best at predicting the sentiment analysis, the constructed models will be assessed on the test set using evaluation metrics such as accuracy, precision, recall and F1 score. Results: From the conducted experimentation the BERT neural network model is the most efficient algorithm in classifying the IMDb movie reviews into positive and negative sentiments. This model achieved the highest accuracy score of 90.67% over the trained data set, followed by the BoW model with an accuracy of 79.15%, whereas the TF-IDF model has 78.98% accuracy. BERT model has the better precision and recall with 0.88 and 0.92 respectively, followed by both BoW and TF-IDF models. The BoW model has a precision and recall of 0.79 and the TF-IDF has a precision of 0.79 and a recall of 0.78. And also the BERT model has the highest F1 score of 0.88, followed by the BoW model having a F1 score of 0.79 whereas, TF-IDF has 0.78. Conclusions: Among the two models evaluated, the lexicon-based approach and the BERT transformer neural network, the BERT neural network is the most efficient, having a good performance score based on the measured performance criteria.

The master's thesis deals with Information Retrieval. It summarizes the knowledge in the field of Information Retrieval theory. Furthermore, the work gives an overview of models used in Information Retrieval, the data and the actual issues and their possible solutions. The practical part of the master's thesis is focused on the implementation of methods of information retrieval in textual data. The last part is dedicated to experiments validating the implementation and its possible improvements.

本文旨在探討關鍵查核事項與會計師事務所特性之關聯性。其中，會計師事務所特性係指其獨立性與專業能力，並分別以任期與產業專家衡量之。 　　本文實證研究結果如下：其一，主查會計師之任期對關鍵查核事項幾無影響。其二，產業專家會計師事務所與關鍵查核事項之數量及品質僅有部分試驗呈正相關。其三，產業專家主查會計師其對關鍵查核事項之數量及品質均有正面影響。 　　在增額測試的部分，結果如下：其一，對產業專家主查會計師而言，任期對關鍵查核事項有正面效果。其二，會計師事務所與主查會計師俱為產業專家的會計師對關鍵查核事項之正面影響力大於僅有會計師事務所為產業專家的會計師。最後，會計師的專業能力使其更有能力以簡明扼要的文字呈現關鍵查核事項。 / The objective of this thesis is to investigate the relationship between key audit matters (KAMs) and audit firm characteristics. In this study, audit firm characteristics are focused on their independence and ability, and are measured by tenure and industry expertise, respectively. The empirical results can be summarized as follows. Firstly, lead partners tenure has little effect on KAMs. Secondly, partial evidence is found on the association between firm-level industry specialist auditors and KAMs. Thirdly, partner-level industry specialist auditors have positive effects on both the quantity and quality of KAMs. In further examinations, the results are as follows. Firstly, tenure has positive effects on KAMs when the auditors are partner-level industry specialist. Secondly, industry experts at both firm- and partner-levels have stronger positive effects on KAMs then industry experts at firm-level alone. Lastly, auditors’ capacity allows them to present KAMs more concisely.

Energy forms a vital input and critical infrastructure for the economic development of countries and for improving the quality of life of people. Energy is utilized in society through the operation of large socio-technical systems called energy systems. In a growing world, as the focus shifts to better access and use of modern energy sources, there is a rising demand for energy. However, certain externalities result in this demand not being met adequately, especially in developing countries. This constitutes the energy demand – supply matching problem. Load shedding is a response used by distribution utilities in developing countries, to deal with the energy demand – supply problem in the short term and to secure the grid. This response impacts the activities of consumers and entails economic losses. Given this scenario, demand – supply matching becomes a crucial decision making activity. Traditionally demand – supply matching has been carried out by increasing supply centrally in the long term or reducing demand centrally in the short term. Literature shows that these options have not been very effective in solving the demand-supply problem. Gaps in literature also show that the need of the hour is the design of alternate solutions which are tailored to a nation's specific energy service needs in a sustainable way. Microgrids using renewable and clean energy resources and demand side management can be suitable decentralized alternatives to augment the centralized grid based systems and enable demand – supply matching at a local community level. The central research question posed by this thesis is: “How can we reduce the demand – supply gap existing in a community, due to grid insufficiency, using locally available resources and the grid in an optimal way; and thereby facilitate microgrid implementation?” The overall aim of this dissertation is to solve the energy demand – supply matching problem at the community level. It is known that decisions for the creation of energy systems are influenced by several factors. This study focuses on those factors which policy-makers and stakeholders can influence. It proposes an integrated decision framework for the creation of community microgrids. The study looks at several different dimensions of the existing demand – supply problem in a holistic way. The research objectives of this study are: 1. To develop an integrated decision framework that solves the demand – supply matching problem at a community level. 2. To decompose the consumption patterns of the community into end-uses. solar thermal, solar lighting and solar pumps and a combination of these at different capacities. The options feasible for medium income consumers are solar thermal, solar pumps, municipal waste based systems and a combination of these. The options for high income consumers are municipal waste based CHP systems, solar thermal and solar pumps. Residential consumers living in multi-storied buildings also have the options of CHP, micro wind and solar. For cooking, LPG is the single most effective alternative. 3. To identify the ‗best fitting‘ distributed energy system (microgrid), based on the end-use consumption patterns of the community and locally available clean and renewable energy resources, for matching demand – supply at the community level. 4. To facilitate the implementation of microgrids by * Contextualizing the demand – supply matching problem to consider the local social and political environment or landscape, * Studying the economic impact of load shedding and incorporating it into the demand-supply matching problem, and * Presenting multiple decision scenarios, addressing the needs of different stakeholders, to enable dialogue and participative decision making. A multi-stage Integrated Decision Framework (IDF) is developed to solve the demand - supply matching problem in a sequential manner. The first stage in the IDF towards solving the problem is the identification and estimation of the energy needs / end-uses of consumers in a community. This process is called End-use Demand Decomposition (EUDD) and is accomplished by an empirical estimation of consumer electricity demand based on structural and socio-economic factors. An algorithm/ heuristic is also presented to decompose this demand into its constituent end-uses at the community level for the purpose of identifying suitable and optimal alternatives/ augments to grid based electricity. The second stage in the framework is Best Fit DES. This stage involves identifying the “best-fit‘ distributed energy system (microgrid) for the community that optimally matches the energy demand with available forms of supply and provides a schedule for the operation of these various supply options to maximize stakeholder utility. It provides the decision makers with a methodology for identifying the optimal distributed energy resource (DER) mix, capacity and annual operational schedule that “best fits” the given end-use demand profile of consumers in a community and under the constraints of that community such that it meets the needs of the stakeholders. The optimization technique developed is a Mixed Integer Linear Program and is a modification of the DER-CAM™ (Distributed Energy Resources Customer Adoption Model), which is developed by the Environmental Energy Technologies Division, Lawrence Berkeley National Laboratory using the GAMS platform. The third stage is the Community Microgrid Implementation (CMI) stage. The CMI stage of IDF includes three steps. The first one is to contextualize the energy demand and supply for a specific region and the communities within it. This is done by the Energy Landscape Analysis (ELA). The energy landscape analysis attempts to understand the current scenario and develop a baseline for the study. It identifies the potential solutions for the demand - supply problem from a stakeholder perspective. The next step provides a rationale for the creation of community level decentralized energy systems and microgrids from a sustainability perspective. This is done by presenting a theoretical model for outage costs (or load shedding), empirically substantiating it and providing a simulation model to demonstrate the viability for distributed energy systems. Outage cost or the cost of non supply is a variable that can be used to determine the need for alternate systems in the absence/ unavailability of the grid. The final step in the CMI stage is to provide a scenario analysis for the implementation of community microgrids. The scenario analysis step in the framework enlightens decision makers about the baselines and thresholds for the solutions obtained in the “best fit‘ analysis. The first two stages of IDF, EUDD and Best Fit DES, address the problem from a bottom-up perspective. The solution obtained from these stages constitutes the optimal solution from a technical perspective. The third stage CMI is a top-down approach to the problem, which assesses the social and policy parameters. This stage provides a set of satisficing solutions/ scenarios to enable a dialogue between stakeholders to facilitate implementation of microgrids. Thus, IDF follows a hybrid approach to problem solving. The proposed IDF is then used to demonstrate the choice of microgrids for residential communities. In particular, the framework is demonstrated for a typical residential community, Vijayanagar, situated in Bangalore and the findings presented. The End-use Demand Decomposition (EUDD) stage provides the decision makers with a methodology for estimating consumer demand given their socio-economic status, fuel choice and appliance profiles. This is done by the means of a statistical analysis. For this a primary survey of 375 residential households belonging to the LT2a category of BESCOM (Bangalore Electricity Supply Company) was conducted in the Bangalore metropolitan area. The results of the current study show that consumer demand is a function of the variables family income, refrigeration, entertainment, water heating, family size, space cooling, gas use, wood use, kerosene use and space heating. The final regression model (with these variables) can effectively predict up to 60% of the variation in the electricity consumption of a household ln(ElecConsumption) = 0.2880.396*ln(Income)+0.2 66*Refri geration+ 0.708*Entertainment+0.334*WaterHeating+0.047*FamSize+ 0243*SpaceCooling.+580*GasUse+0.421*WoodUse–0.159*KeroseneUse+ 0.568*SpaceHeating ln(ElecConsumption) = 0.406*ln(Income)0.168*Ref rigeration+0.139*Entertainment+ 0.213*WaterHeating+0.114*FamSize+0.121*SpacCooling+0.171*GasUse+ 0.115*WoodUse–0.094*KeroseneUse+0.075*SpaceHeating   The next step of EUDD is to break up the demand into its constituent end-uses. The third step involves aggregating the end-uses at the community level. These two steps are to be performed using a heuristic. The Best Fit DES stage of IDF is demonstrated with data from an urban community in Bangalore. This community is located in an area called Vijayanagar in Bangalore city. Vijayanagar is a mainly a residential area with some pockets of mixed use. Since grid availability is the constraining parameter that yields varying energy availability, this constraint is taken as the criteria for evaluation of the model. The Best Fit DES model is run for different values of the grid availability parameter to study the changes in outputs obtained in DER mix, schedules and overall cost of the system and the results are tabulated. Sensitivity analysis is also performed to study the effect of changing load, price options, fuel costs and technology parameters. The results obtained from the BEST Fit DES model for Vijayanagar illustrate that microgrids and DERs can be a suitable alternative for meeting the demand – supply gap locally. The cost of implementing DERs is the optimal solution. The savings obtained from this option however is less than 1% than the base case due to the subsidized price of grid based electricity. The corresponding costs for different hours of grid availability are higher than the base case, but this is offset by the increased efficiency of the overall system and improved reliability that is obtained in the community due to availability of power 24/7 regardless of the availability of grid based power. If the price of grid power is changed to reflect the true price of electricity, it is shown that DERs continue to be the optimal solution. Also the combination of DERs chosen change with the different levels of non-supply from the grid. For the study community, Vijayanagar, Bangalore, the DERs chosen on the basis of resource availability are mainly discrete DERs. The DERs chosen are the LPG based CHP systems which run as base and intermediate generating systems. The capacity of the discrete DERs selected, depend on the end-use load of the community. Biomass based CHP systems are not chosen by the model as this technology has not reached maturity in an urban setup. Wind and hydro based systems are not selected as these resources are not available in Vijayanagar. The CMI stage of IDF demonstrates the top-down approach to the demand-supply matching problem. For the Energy Landscape Analysis (ELA), Bangalore metropolis was chosen in the study for the purpose of demonstration of the IDF framework. Bangalore consumes 25% of the state electricity supply and its per capita consumption at 1560kWh is higher than the state average of 1230kWh and is 250% more than the Indian average of 612kWh. A stakeholder workshop was conducted to ascertain the business value for clean and renewable energy technologies. From the workshop it was established that significant peak power savings could be obtained with even low penetrations of distributed energy technologies in Bangalore. The feasible options chosen by stakeholders for low income consumers are The second step of CMI is finding an economic rationale for the implementation of community microgrids. It is hypothesized that the ‘The cost of non-supply follows an s-shaped curve similar to a growth curve.’ It is moderated by the consumer income, consumer utility, and time duration of the load shedding. A pre and post event primary survey was conducted to analyze the difference in the pattern of consumer behaviour before and after the implementation of a severe load shedding program by BESCOM during 2009-10. Data was collected from 113 households during February 2009 and July 2010. The analysis proves that there is indeed a significant difference in the number of uninterrupted power systems (inverters) possessed by households. This could be attributed mainly to the power situation in Karnataka during the same period. The data also confirms the nature of the cost of non-supply curve. The third step in CMI is scenario analysis. Four categories of scenarios are developed based on potential interventions. These are business-as-usual, demand side, supply side and demand-supply side. About 21 scenarios are identified and their results compared. Comparing the four categories of scenarios, it is shown that business-as-usual scenarios may result in exacerbation of the demand-supply gap. Demand side interventions result in savings in the total costs for the community, but cannot aid communities with load shedding. Supply side interventions increase the reliability of the energy system for a small additional cost and communities have the opportunity to even meet their energy needs independent of the grid. The combination of both demand and supply side interventions are the best solution alternative for communities, as they enable communities to meet their energy needs 24/7 in a reliable manner and also do it at a lower cost. With an interactive microgrid implementation, communities have the added opportunity to sell back power to the grid for a profit. The thesis concludes with a discussion of the potential use of IDF in policy making, the potential barriers to implementation and minimization strategies. It presents policy recommendations based on the framework developed and the results obtained.

The master's thesis deals with automatic classifi cation of text document. It explains basic terms and problems of text mining. The thesis explains term clustering and shows some basic clustering algoritms. The thesis also shows some methods of classi fication and deals with matrix regression closely. Application using matrix regression for classifi cation was designed and developed. Experiments were focused on normalization and thresholding.

The topic of this thesis is the design and realization of a modular Data concentrator for various types of measurements in diverse conditions. The device conception, basic principles, design and functionality are described herein. Also, the lightweight protocol design for radio communication, standardized XML format for data storage and cloud usage are described. The electrical design of the device as well as the design of the printed circuit board was made using the Eagle Autodesk electronic design automation software (Eagle Autodesk EDA). The control software written in C/C++ for a target microcontroller (ESP32-WROOM-32) is based on a FreeRTOS platform and ESP-IDF framework. An IDE for managing this software is Visual Studio Code with PlatformIO extension. Selected and used Cloud Platform is ThingSpeak from Mathworks, which uses certain components from Matlab platform.