This dissertation is focused on the design and evaluation of machine learning algorithms for modeling information precursors for use in event modeling and forecasting. Given an online stream of information (e.g., news articles, social media postings), how can we model and understand how events unfold, how they influence each other, and how they can act as determinants of future events? First, we study information reciprocity in joint news and social media streams to capture how events evolve. We present an online story chaining algorithm which links related news articles together in a low complexity manner and a mechanism to classify the interaction between a news article and social media (Twitter) activity into four categories. This is followed by identification of major information sources for a given story chain based on the interaction states of news and Twitter. We demonstrate through this study that Twitter as a social network platform serves as a fast way to draw attention from the public to many social events such as sports, whereas news media is quicker to report events regarding political, economical, and business issues. In the second problem we focus on forecasting and understanding large-scale societal events from open source datasets. Our goal here is to develop algorithms that can automatically reconstruct precursors to societal events. We develop a nested framework involving multi-instance learning for mining precursors by harnessing temporal constraints. We evaluate the proposed model for various event categories in multiple geo-locations with comprehensive experiments. Next, to reinforce the fact that events are typically inter-connected and influenced by events in other locations, we develop an approach that creates personalized models for exploring spatio-temporal event correlations; this approach also helps tackle data/label sparsity problems across geolocations. Finally, this dissertation demonstrates how our algorithms can be used to study key characteristics of mass events such as protests. Some mass gatherings run the risk of turning violent, causing damage to both property and people. We propose a tailored solution for uncovering triggers from both news media and social media for violent event analysis. This work was partially supported by the Intelligence Advanced Research Projects Activity (IARPA) via Department of Interior National Business Center (DoI/NBC) contract number D12PC000337, the Office of Naval Research under contract N00014-16-C-1054, and the U.S. Department of Homeland Security under Grant Award Number 2017-ST-061-CINA01. The US Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright annotation thereon. The views and conclusions contained herein are those of the author and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of NSF, IARPA, DoI/NBC, or the US Government. / Ph. D. / Today, massive open source information is widely available through news and social media, but analyzing this information is a complex task. It is imperative to develop algorithms that can automatically reconstruct the clues to societal events that are reported in news or social media. The focus of this dissertation is on simultaneously uncovering precursors to societal events and using such precursors to forecast upcoming events. We develop various machine learning algorithms that can model event-related data and determine the key happenings prior to an event that have the greatest predictability to such events in the future. We use our algorithms to understand the nature of precursors to civil unrest events (protests, strikes, and ‘occupy’ events) and why some of these events turn violent.

No description available.

No description available.

Finding frequent patterns plays a vital role in many analytics tasks such as finding itemsets, associations, correlations, and sequences. In recent decades, spatiotemporal frequent pattern mining has emerged with the main goal focused on developing data-driven analysis frameworks for understanding underlying spatial and temporal characteristics in massive datasets. In this thesis, we will focus on discovering spatiotemporal event sequences from large-scale region trajectory datasetes with event annotations. Spatiotemporal event sequences are the series of event types whose trajectory-based instances follow each other in spatiotemporal context. We introduce new data models for storing and processing evolving region trajectories, provide a novel framework for modeling spatiotemporal follow relationships, and present novel spatiotemporal event sequence mining algorithms.

La commande événementielle offre une alternative prometteuse à la commande périodique classique, qui est considérée comme peu économe vis-à-vis des ressources. Contrairement à la commande classique, la commande événementielle propose de passer d'une loi de commande en temps continu à une loi de commande numérique à travers un échantillonnage non-uniforme. Dans ce cas, une nouvelle valeur de la loi de commande n'est calculée que lorsque la réponse du système est inadmissible. En revanche, la valeur de la loi de commande est maintenue constante si la réponse est satisfaisante. Dans cette thèse, nous explorons des moyens de réduire le nombre de mise à jour de la loi de commande, et de rallonger les intervalles de temps entre les mises à jour.Dans le cas des systèmes linéaires, nous présentons un algorithme de stabilisation dans lequel nous relaxons les conditions de stabilité sur la fonction de Lyapunov du système. Pour induire moins d'échantillons, on requiert uniquement que cette fonction soit maintenue sous un seuil décroissant. Le calcul des paramètres optimaux du seuil est transformé en problème de valeurs propres généralisées. Ensuite, cette approche est étendue aux systèmes dits switched, et une version self-triggered est proposée. Nous traitons également le problème de suivi de trajectoire par une commande événementielle. Enfin, dans le cas du contrôle des systèmes non-linéaires, nous proposons d'utiliser une analyse de contraction des trajectoires, et ce à cause de la difficulté de trouver une fonction de Lyapunov pour ces systèmes. / Event-triggered control offers a promising alternative to the classical, resource-consuming, periodic control. It suggests to replace the periodic, high frequency sampling used in the continuous-to-discrete transformations of control signals with aperiodic sampling. A new value of the event-triggered control law is computed only when the system's response is unsatisfactory. The control value is kept constant otherwise. In this thesis, we explore ways to induce fewer updates, and to have longer intervals between two samples. We also seek to make the algorithms that we design more detailed, by describing how to choose or compute the optimal parameters.In the linear case, we present a stabilizing algorithm in which we relax the stability conditions on the system's Lyapunov function to produce fewer, sparser updates of the control law. Stability is ensured by maintaining the Lyapunov function below a certain decreasing threshold. The optimal threshold function is derived by solving a maximum generalized eigenvalue problem. This approach is then extended to switched linear systems. We also present a self-triggered version of this algorithm using Newton methods for optimization and root-finding. The reference tracking problem is treated in the event-triggered control framework as well. Finally, in the nonlinear case, due to the difficulty of finding a Lyapunov function, we explore the use of contraction analysis. This approach allows us to describe the nonlinear event-triggered control algorithm more thoroughly than if we had used Lyapunov techniques.

"Complex Event Processing (CEP) has become increasingly important for tracking and monitoring anomalies and trends in event streams emitted from business processes such as supply chain management to online stores in e-commerce. These monitoring applications submit complex event queries to track sequences of events that match a given pattern. While the state-of-the-art CEP systems mostly focus on the execution of flat sequence queries, we instead support the execution of nested CEP queries specified by the (NEsted Event Language) NEEL. However the iterative execution often results in the repeated recomputation of similar or even identical results for nested sub- expressions as the window slides over the event stream. This work proposes to optimize NEEL execution performance by caching intermediate results. In particular a method of applying selective caching of intermediate results called Continuous Sliding Caching technique has been designed. Then a further optimization of the previous technique which we call the Semantic Caching and the Continuous Semantic Caching have been proposed. Techniques for incrementally loading, purging and exploiting the cache content are described. Our experimental study using real- world stock trades evaluates the performance of our proposed caching strategies for different query types."

V práci je popsán teoretický postup při pořádání eventů, který vychází ze zásad projektového řízení. Tyto postupy jsou také aplikovány v praktické realizaci eventu. Následuje výzkum aktuálních trendů v event managementu a výzkum cílových skupin s pomocí dat MML/TGI.

This thesis investigates the impact of terrorist attacks on equity financial markets. It employs traditional event study approaches to identify and measure stock market reactions to terrorist attacks in New York on September 11, 2001, and subsequent terrorist attacks in Madrid, London and Bali. Three studies are presented. The first study investigates the impact of September 11 on the tenant firms within the World Trade Centre. The second study investigates industry effects following the Madrid and London bombings. The third study undertakes a sensitivity analysis to different event study techniques over the various terrorist attacks. The results from the three studies suggest that equity markets can remain efficient in the wake of terrorist events. Terrorist events can trigger large abnormal movement in both equity prices and volume traded. These price and volume effects are influenced by industry effects. Terrorism has a differential impact on stock markets and industry portfolios within stock markets. The detailed analysis presented in this thesis can be used to exploit that industry effect and can be employed to guide diversification strategies that could minimize terrorist risk through industry diversification. The thesis has also evaluated alternative event study methods and produced a critical analysis of event study methodology. It shows clearly that methodological choices can and do significantly influence results. The thesis contributes to eliminating some uncertainty about the markets response to terrorist events, and identifies opportunities for reducing terrorist risk in stock markets.

<p>Uppsatsens titel ”Work is theatre and every business is a stage” – en studie i möjliggörandet av relationer med hjälp av event marketing</p><p>Seminariedatum 2008-06-05</p><p>Ämne/kurs Kandidatuppsats i marknadsföring, 15 p</p><p>Författare Sophie Ohlsson och Dina Sabrkesh</p><p>Handledare Ingemar Haglund</p><p>Nyckelord Event marketing, relationer, relationsmarknadsföring, evenemang,</p><p>event.</p><p>Problemområde Konsumenter idag vill inte dagligen pumpas med nya budskap utan de har blivit mer medvetna och söker efter ett värde från företaget och dess produkter utöver det brukliga. För att tillgodose detta har företagen börjat inrikta sin marknadsföring alltmer mot att försöka skapa interaktion, gemenskap och relation med sin målgrupp och ett effektivt sätt att göra det på är just genom event marketing. Genom dessa möten kan företag skapa och bibehålla relationen till sina kunder samt öka medvetenheten kring varumärket. Ett starkt varumärke bidrar till att möjliggöra relationer då ett välkänt varumärke skapar trygghet hos konsumenten.</p><p>Syfte Syftet med uppsatsen är att beskriva processen kring planerandet och genomförandet av ett event. Detta för att på så sätt närmare kunna beskriva hur man med hjälp av event marketing kan möjliggöra relationer bland nuvarande och före detta studenter inom ett utbildningsprogram.</p><p>Metod För att uppfylla syftet med uppsatsen har ett enskilt fall studerats. Fallet, som är ett event, var inte bara något som observerades utan också planerades och genomfördes. För att tränga igenom ytan på problemet och skapa förståelse för vad som specifikt sker i en viss situation har uppsatsen utgått från en fallstudieansats. En fallstudie är av kvalitativ karaktär där ett eller flera fall studeras under ett tillfälle eller över tiden. Den kvalitativa karaktären hjälper till att beskriva den sociala verkligheten vad gäller händelser, människor och handlingar.</p><p>Slutsats Vid planering och genomförande av ett event är det av stor betydelse att planeringsprocessen genomförs noggrant och att arrangören är strukturerad och välorganiserad. För att möjliggöra relationer vid eventet är det viktigt att det finns en röd tråd mellan upplevelsen, mötet och budskapet som tilltalar den utvalda målgruppen. Vidare bör deltagarna ha en gemensam nämnare förutom eventet då det underlättar konversationer. Det är också viktigt att arrangören och deltagarna har ett stort engagemang. I de fall där relationer skapas eller bibehålls kan det leda till att programmet stärks. Relationer med näringslivet kan innebära praktikplatser vilket är ett attraktivt marknadsföringsargument vid rekrytering av nya studenter. Vidare kan relationer leda till färre avhopp samt en ökad spridning av positiv word of mouth.</p>

<p>The society of today is overflowed with information and commercials. There is a thick media noise that the marketers finds hard to penetrate. Bearing this in mind, we can see that Event Marketing, witch is marketing through events, has grown to be a popular form of communication as it has an ability to work its way through the existing media noise. Through events a company can strengthen its brand and image as well as form a lasting relationship with its consumers. What other functions does Event Marketing possess? In what context is Event Marketing an appropriate marketing strategy to use? These are two questions that we aim to answer through this essay.</p><p>To be able to answer these questions we interviewed the marketing manager of two different companies, L´oréal and TeliaSonera. One of them sells hair care products and the other one sells telecommunications services.</p><p>We also interviewed two different Event Marketing bureaus, Subito and BSP-marketing, to be able to create a deeper understanding of the vast area of Event Marketing.</p><p>At last we interviewed two scientists within the area of marketing, to gain a wider objective of the subject Event Marketing. This essay is based on qualitative, case based research.</p><p>We analysed the empirics that we sampled from our informants through the different theories that we hade chosen to apply on this study. The results that we acquired helped us to answer our questions about Event Marketing.</p><p>We found out that Event Marketing, as stated, can help a company to produce lasting relations. Through an event the consumer gets an opportunity to actively participate which is creating a larger engagement from the consumers. This engagement helps the target audience to remember the brand when it is time for them to make a buying decision.</p><p>By letting the consumer themselves choose what event to attend they will also find the message that the company is trying to convey more trustworthy.</p><p>Through Event Marketing the companies are given a chance to physically meet and influence their target audiences. This physical contact between the consumer and the company helps the company to be heard and seen in the compact media noise that surrounds us today.</p><p>We also found out that it is possible to use Event Marketing as a strategy for all companies and on all target audience. It is, however, important that the event created has a clear goal and purpose.</p>