In the field of data mining, transfer learning is the method of transferring knowledge from one domain into another. Using reviews from prisjakt.se, a Swedish price comparison site, and hotels.com this work investigate how the similarities between domains affect the results of transfer learning when using recurrent neural networks. We test several different domains with different characteristics, e.g. size and lexical similarity. In this work only relatively similar domains were used, the same target function was sought and all reviews were in Swedish. Regardless, the results are conclusive; transfer learning is often beneficial, but is highly dependent on the features of the domains and how they compare with each other’s.

Detta är en studie om instrumental låtskrivning genom adaption – från trummor till gitarr. Mitt syfte med studien var att undersöka hur trumkomp och rudiments kan tolkas på gitarren och användas som verktyg för låtskrivning. Genom att använda trumkomp och rudiments för att skapa gitarrkomp samt rudiments som verktyg och inspiration till melodi-skrivande på gitarr skapades fyra instrumentala låtar. I uppsatsen beskrivs hur denna tolkning gick till och hur adaptionen kunde vara mer eller mindre kreativ beroende på om trumkomp eller rudiments var ursprunget till låten. De fyra låtarna fullbordades med enklare arrangemang och spelades in. I diskussionen resonerar jag kring hur min kreativitet har stimulerats men också hämmats av det här sättet att skriva komponera musik. Slutsatser som jag gör är t.ex att det tillvägagångssätt som beskrivs i uppsatsen kanske bör ses som ett inspirerande första steg i en kreativ process snarare än ett verktyg som ska följas strängt under hela komponerandet av en låt.

Uppsatsen analyserar E.T.A Hoffmanns konstsaga Nötknäpparen, översatt av Ingrid Windisch (1989), mot två andra upplagor av Nötknäpparen, dessa två i bilderboksformat. Den ena bilderboken är återberättad av Anthea Bell men svensk översatt av Gun-Britt Sundström (1987) och den senaste upplagan av Nötknäpparen är återberättad av Rachel Elliot men svenskt översatt av Annika Meijer (2016). Syftet med denna analys är att främst undersöka vilka adaptioner som gjorts ifrån originalberättelsen till de två bilderboksupplagorna. Fokus kommer vara på karaktärer och intrigen i samtliga böcker med stort fokus på vad som bland annat adapterats från den översatta originalberättelsens otäcka innehåll till bilderböckerna. Uppsatsen utgår ifrån en narratologisk och komparativ analys där även begrepp som konstsaga, bilderbok, karaktär, intrig samt adaption kommer ligga till grund för att besvara syfte och frågeställningar. Analysen visar att mycket är lika i de karakteristiska dragen men att det betonas olika mycket i varje bok. Intrigen i Hoffmanns berättelse och bilderboken från 1987 är till stor del densamma dock är bilderboken förkortad och innehåller illustrationer. Bilderboken från 2016 innehåller till viss del samma delar i intrigen men har exempelvis tagit bort delen om prinsessan Pirlipat och nöten Krakatuk. Bilderboken (2016) är mer anpassad efter yngre barn i åldern 3-6 år vilket märks både i karaktärer och i intrig. Det framkommer även i analysen att Hoffmanns berättelse har ett mer detaljerat skrämmande innehåll än övriga bilderböcker även om bilderboken (1987) är relativt mycket mer skrämmande än bilderboken (2016) både i text och illustrationer.

Som filmskapare i en reklamvärld där utvecklingen ständigt går mot att göra kortare reklamfilmer, kände vi behov av att analysera vad som vinns och förloras i processen. Uppsatsen berör hur en längre välgörenhetsreklamfilm påverkas av att kortas ned i två olika tidsramar. Syftet var att hitta mönster som kan bidra till en ökad förståelse i ämnet för filmskapare. Uppsatsen analyserar de olika versionernas berättarstruktur, tematik, retorik, empati- och sympatiförutsättningar för att sedan diskutera kring hur effektiva de är i förhållande till filmens syfte. Slutsatsen visade att versionerna har olika egenskaper och därmed olika fördelar och nackdelar. För filmskapare innebär det att med noggrannhet och god analysförmåga kunna välja ut bilder i adaptionsprocessen som ger det mest önskvärda resultatet, för att vara säker på att det uttänkta syftet nås fram till åskådaren.

In this thesis, we investigate the ergodic capacity under several power adaption schemes, including optimal power and rate algorithm (OPRA), optimal rate algo rithm (ORA), channel inversion (CI), and truncated channel inversion (TCI), over fluctuating two-ray (FTR) fading channels and Fisher-Snedecor F fading channels. After some mathematical manipulations, the exact expressions for the EC under those power adaption schemes are derived. To simplify the expressions and also get some insights from the analysis, the corresponding asymptotic expressions for the EC are also derived in order to show the slope and power offset of the EC in the high signal-to-noise ratio (SNR) region. These two metrics, i.e., slope and power offset, govern the EC behaviour in the high SNR region. Specifically, from the derived asymptotic expressions, we find that the slope of the EC of OPRA and ORA over FTR fading channels is always unity with respect to the average SNR in the log-scale in high SNRs, while the asymptotic EC of the TCI method is not a line function in the log-scale. For the Fisher-Snedecor F fading channel, the slope of asymptotic EC under OPRA, ORA, and CI (m > 1) schemes is unity in the log-scale, where m is the fading parameter. The slope of the TCI method depends on m, i.e., unity for m > 1 and m for m > 1, while the asymptotic EC of TCI is not a line function for m = 1. Finally, Monte-Carlo simulations are used to demonstrate the correctness of the derived expressions.

This study involved a 28-week school-based exercise trial of single-leg drop-landing exercise with 42 girls (Tanner stage 1; 6-10 yr old) randomly assigned to control (C), low-drop (LD) or high-drop(HD) exercise groups. The latter two groups performed single-leg drop-landings (3 sessions.wk-1 and 50 landings.session-1) from 14cm and 28cm, respectively using the non-dominant leg. Single-leg peak ground-reaction impact forces (PGRIF) in a sub-sample ranged between 2.5 – 4.4 x body-weight (BW). No differences (p>0.05) among groups at baseline for age, stature, lean tissue mass (LTM - DXA - Lunar 3.6-DPX), leisure time physical activity or average daily calcium intake were detected. No significant within group changes for between leg differences from baseline to post-training and no significant differences among groups at baseline, or in magnitude of change for any of the dominant or non-dominant (loaded) leg bone mineral content (BMC g) measures determined by DXA – loaded leg total - 19.06, 25.5, 25.46 [p=.156], femoral neck - 0.14, 0.11, 0.15 [p=.959], greater trochanter - 0.37, 0.06, 0.26 [p=.733], mid femoral shaft - 3.87, 3.87, 3.42 [p=.677] for the C, LD and HD groups, respectively, after adjusting for the covariates baseline body and fat mass, and change in LTM (ANCOVA) were observed. Similarly, following ANCOVA adjustments no significant differences for changes in calcaneal speed of sound and broadband ultrasound attenuation (CUBA Clinical), DXA derived changes in femoral neck (-0.009, 0.033, -0.009; p=.189) and total MFS (0.029, 0.041, 0.053; p=.447) volumetric BMD (g.cm-3), or MFS cortical volumetric BMD, the latter derived by a new technique combining MRI and DXA were identified. TBBMC changed by 79.6g-C, 100.2g-LD and 91.9g-HD (p=.339). Combining data from both exercise groups to increase statistical power produced similar results. No significant within group changes for between leg differences from baseline to post-training and no significant differences among groups at baseline, or in magnitude of change for any of the dominant or non-dominant (loaded) leg bone geometrical (area cm2) determined by MRI using ANALYZE® software of proximal - 22.18, 12.91, 19.86 [p=.248], mid - 19.83, 15.91, 19.64 [p=.233], or distal - 14.78, 16.07, 13.35 [p=.792], slice cortical area for the C, LD and HD groups, respectively, after adjusting for the covariates baseline body and fat mass, and change in LTM (ANCOVA) were detected. Similarly there were no significant biomechanical cross sectional moment of inertia (CSMI cm4) changes determined by Scion Image® (Frederick, Maryland: Version-Beta 3B) and a custom macro program of proximal - 896, 815, 649 [p=.415], mid - 1054, 806, 1087 [p=.471], or distal - 1197, 1079, 966 [p=.606], slice CSMI for the C, LD and HD groups, respectively after adjusting for the same covariates. In contrast to some recent reports, our findings suggest that strictly controlled uni-modal; uni-directional single-leg drop-landing exercises involving low-moderate peak ground-reaction impact forces are not osteogenic in the developing prepubertal female skeleton.

No description available.

<p>Climate change caused by green house gas emissions, mainly carbon dioxide, is today’s most debated environmental issue. The United Nations Framework Convention on Climate Change, with the legally binding Kyoto protocol, is the emission regulatory framework. Tanzania has ratified both the conventions supporting carbon reductions.</p><p>Tanzania has a very varied climate with two rainfall regimes dominating the country, bimodal and unimodal. Scientific literature predicts a 2,2-4°C average increase in temperature for Tanzania, all studies also predict a higher increase in the cooler period and lower in the warm period. Rainfall predictions are less certain but in general a 10 % average increase is expected but the distribution uneven, both between rain periods and geographically. Tanzanian reports estimate a 5-45 % increase in rainfall in bimodal regions and a 5-15 % decrease in unimodal regions. The distribution of increase in bimodal regions will be uneven, with an increase in the long rain period and a decrease in the short rain period.</p><p>Agriculture is indisputably the most important source of income in Tanzania standing for 80% of employment and 50% of GNP. Climate change will therefore inevitably affect the economy and livelihood of people. Developing countries are also more vulnerable due to lower adaptation capacity. Effects on Tanzania are predicted to be both positive and negative. Maize production, the major staple, is predicted an average 33% decline while cash crops like coffee and cotton are predicted to increase. To be able to adapt there first has to be a perception of need to adapt, adaption strategies then have to be developed and barriers worked through. The purpose of this thesis is to investigate the level of awareness, strategies and barriers existing for adaption to climate change on national, regional and local level, Tanzania nationally and Babati regionally and locally. The research questions are; How do people perceive climate change? What are the strategies for adapting to climate change? What are the barriers for adaption to climate change? At national level policy framework and strategies were used to answer the research questions. At regional and local level a field study was conducted in Babati to answer the research questions at these levels.</p><p>The study showed that Tanzania nationally has declared their conviction in climate change and state that the issue has to be addressed. Agriculture is identified as one of the most vulnerable areas however climate change is not mainstreamed into agricultural and environmental policy framework. In Babati district no policy was found on climate change and official perception varied. Concerns were related to timing and amount of rainfall, the results were the same for farmers. Global climate change was also known for both studied groups and existed as a parallel truth with the local reason for changes. A number of adaption strategies are also identified nationally, both used and potential, where small scale irrigation is the primary adaptive step. Switching to draught resistant crops is also prioritized in the North eastern region. In Babati adaption strategies were promoted, even though there was not a general policy or perception, by officials to switch to short-term crops and planting of trees. Switching cops was also the most commonly used strategy by farmers along with traditional diversification. Nationally a large number of barriers are also identified including, lack of funding, poverty, HIV, lack of infrastructure and analytic capacity. Officials in Babati also mentioned the lack of money, deforestation, lack of clouds, education, irrigation and seeds. The farmers in Babati were not so clear about what they needed to adapt, irrigation, livestock backup, diversification and switching crops were mentioned, not differing much from used strategies.</p>

<p>Scania, the southernmost region in Sweden, is facing a changing climate, and even if we limitcarbon emissions the climate will continue to change. Increasing amounts of precipitation, butalso an increased frequency of heavy rainfall, drought and increased temperature are examplesof some consequences, but it is uncertain how these will affect water supplies. To understandthe whole picture of the climate adaption, we have to reach out with information, knowledge,advice and together during training courses, seminars and workshops incorporate climateadaptation efforts. The basis of this project is a telephone interview with Scania´s 33municipalities over with the contingency contained in a changing climate. Work on coastalerosion and flooding risks to coastal communities is fairly well inventoried, but equalattention has not been given to threats and risks that exist to our water, especially drinkingwater. Many times an alternative water supply does not exist if current supplies are impacted.Municipalities must plan for our water resources in the long term, because safe and securedrinking water supplies are needed today and in the future. An important part of climateadaptation efforts is the establishment of water supply plans, in order to ensure the long-termsupply of drinking water. The plan should include two main parts; the first part is an inventoryof water resources and water conservation for current and future drinking water supply andthe second part is action-oriented and based on the first part with an inventory of the potentialimpacts and threats to these resources.</p>

Scania, the southernmost region in Sweden, is facing a changing climate, and even if we limitcarbon emissions the climate will continue to change. Increasing amounts of precipitation, butalso an increased frequency of heavy rainfall, drought and increased temperature are examplesof some consequences, but it is uncertain how these will affect water supplies. To understandthe whole picture of the climate adaption, we have to reach out with information, knowledge,advice and together during training courses, seminars and workshops incorporate climateadaptation efforts. The basis of this project is a telephone interview with Scania´s 33municipalities over with the contingency contained in a changing climate. Work on coastalerosion and flooding risks to coastal communities is fairly well inventoried, but equalattention has not been given to threats and risks that exist to our water, especially drinkingwater. Many times an alternative water supply does not exist if current supplies are impacted.Municipalities must plan for our water resources in the long term, because safe and securedrinking water supplies are needed today and in the future. An important part of climateadaptation efforts is the establishment of water supply plans, in order to ensure the long-termsupply of drinking water. The plan should include two main parts; the first part is an inventoryof water resources and water conservation for current and future drinking water supply andthe second part is action-oriented and based on the first part with an inventory of the potentialimpacts and threats to these resources.