The information age and the information society have become dominant features in the newm illennium.H owever,t heset ermsa reo ften referredt o with the youngerg enerations in mind,n eglectingth e older andm oree xperiencemd emberso f our society. This thesis focuses on the 'Grey Market' (travellers over the age of 50) who use the Internet on a regular basis - therefore also referred to as 'Silver Surfers' - and in particularf or their travel and tourism decision-makingI.t aims to identify experiences andp rocesseosf travel decision-makinga,n alyseth e impacto n the useo f the Interneta s an informations earcha s well as evaluateth e effectivenesos f the Interneti n providing informationf or particulara ndn ot mainstreamm arkets egments. The methodologye mployedi n this particularp iece of researchb uilds on positivisma s most consumerb ehaviourt heoriesd o; howevera more inductivea pproachw as taken. While relying on existingt heoriesn ewera nd lessw ell testedm ethodso f datac ollection were put to use.T he methodsw ere triangulatedu, tilising bothq uantitativea ndq ualitative research methods which complement each other in the results. An initial pilot study questionnairwe asf ollowedu p with semi-structureidn -depthi nterviewsw hich thenl edt o the completiono f the final survey,t hat was administeredb y 'e-surveying'u sing both conveniencea nd snowballs amplinga nd resultedi n 517 valid responsesfr om 'Silver Surfers' around the United Kingdom. Main findings of this thesiss how a distinct patterno f behaviourin the travel decisionmaking process of this particular market segment as well as what kind of information they were researchingo n the Internet.M ost importantly,t he respondentdso not consider themselvesto o different from other (younger)a geg roupsa nde vent houghs omeo f their informationr equirementsa re distinctive,t hey do not want to be consideredju st as 'the older consumers'.

Online decision making is a kind of machine learning problems where decisions are made in a sequential manner so as to accumulate as many rewards as possible. Typical examples include multi-armed bandit (MAB) problems where an agent needs to decide which arm to pull in each round, and network routing problems where each router needs to decide the next hop for each packet. Thompson sampling (TS) is an efficient and effective algorithm for online decision making problems. Although TS has been proposed for a long time, it was not until recent years that the theoretical guarantees for TS in the standard MAB were given. In this thesis, we first analyze the performance of TS both theoretically and practically in a special MAB called combinatorial MAB with sleeping arms and long-term fairness constraints (CSMAB-F). Then, we apply TS to a novel reactive network routing problem, called \emph{opportunistic routing without link metrics known a priori}, and use the proof techniques we developed for CSMAB-F to analyze the performance. / Graduate

<p dir="ltr">We propose a fully online procedure to conduct statistical inference with adaptively collected data. The low-rank structure of the model parameter and the adaptivity nature of the data collection process make this task challenging: standard low-rank estimators are biased and cannot be obtained in a sequential manner while existing inference approaches in sequential decision-making algorithms fail to account for the low-rankness and are also biased. To tackle the challenges previously outlined, we first develop an online low-rank estimation process employing Stochastic Gradient Descent with noisy observations. Subsequently, to facilitate statistical inference using the online low-rank estimator, we introduced a novel online debiasing technique designed to address both sources of bias simultaneously. This method yields an unbiased estimator suitable for parameter inference. Finally, we developed an inferential framework capable of establishing an online estimator for performing inference on the optimal policy value. In theory, we establish the asymptotic normality of the proposed online debiased estimators and prove the validity of the constructed confidence intervals for both inference tasks. Our inference results are built upon a newly developed low-rank stochastic gradient descent estimator and its non-asymptotic convergence result, which is also of independent interest.</p>