Reliable forecasting models are needed to achieve the climate related goals in the face of increasing transport demand. Such models can predict the long-term behavioural response to policy interventions, including infrastructure investments, and thus provide valuable pre-dictions for decision makers. Contemporary forecasting models are mainly based on national travel surveys. Unfortunately, the response rates of such surveys have steadily declined, implying that the respondents become less representative of the whole population. A particular weakness is that it is likely that respondents with a high valuation of time are less willing to respond to surveys (because they have less time available for such), and therefore there is a high chance that they are underrepresented among the respondents. The valuation of time plays an important role for the cost benefit analyses of public policies including transport investments, and there is no reliable way of controlling for this uneven sampling of time preferences. Fortunately, there is simultaneously an increase in the number of signals sent between mobile phones and network antennae, and research has now reached the point where it is possible to determine not only the travel destination but also the travel mode based on mobile phone network antennae connections. The aim of this thesis is to investigate if and how mobile phone network data can be used to estimate transportation mode choice demand models that can be used for forecasting and planning. Key challenges with using this data source in the context of mode choice models are identified and met. The identified challenges include uncertainty in the choice variable, the difficulty to distinguish car and bus trips, and the lack of information about the trip purpose. In the first paper we propose three possible model formulations and analyse how the uncertainty in the choice outcome variable would play a role in the different model formulations. We also conclude that it is indeed possible to estimate mode choice demand models based on mobile phone network data, with good results in terms of behavioural interpretability and significance. In the second paper we estimate models using a nested logit structure to account for the difficulty in separating bus and car, and a latent class model specification to meet the challenge of having an unknown trip purpose. / <p><strong>Funding agencies:</strong> The research in this thesis has mainly been funded by the research projects DEMOPAN and DEMOPAN-2 within the research program Transportekonomi at The Swedish Transport Administration.</p>

Malicious users in data networks may form social interactions to create communities in abnormal fashions that deviate from the communication standards of a network. As a community, these users may perform many illegal tasks such as spamming, denial-of-service attacks, spreading confidential information, or sharing illegal contents. They may use different methods to evade existing security systems such as session splicing, polymorphic shell code, changing port numbers, and basic string manipulation. One way to masquerade the traffic is by changing the data rate patterns or use very low (trickle) data rates for communication purposes, the latter is focus of this research. Network administrators consider these communities of users as a serious threat. In this research, we propose a framework that not only detects the abnormal data rate patterns in a stream of traffic by using a type of neural network, Self-organizing Maps (SOM), but also detect and reveal the community structure of these users for further decisions. Through a set of comprehensive simulations, it is shown in this research that the suggested framework is able to detect these malicious user communities with a low false negative rate and false positive rate. We further discuss ways of improving the performance of the neural network by studying the size of SOM's.

<div>Representation learning (RL) for social networks facilitates real-world tasks such as visualization, link prediction and friend recommendation. Many methods have been proposed in this area to learn continuous low-dimensional embedding of nodes, edges or relations in social and information networks. However, most previous network RL methods neglect social signals, such as textual communication between users (nodes). Unlike more typical binary features on edges, such as post likes and retweet actions, social signals are more varied and contain ambiguous information. This makes it more challenging to incorporate them into RL methods, but the ability to quantify social signals should allow RL methods to better capture the implicit relationships among real people in social networks. Second, most previous work in network RL has focused on learning from homogeneous networks (i.e., single type of node, edge, role, and direction) and thus, most existing RL methods cannot capture the heterogeneous nature of relationships in social networks. Based on these identified gaps, this thesis aims to study the feasibility of incorporating heterogeneous information, e.g., texts, attributes, multiple relations and edge types (directions), to learn more accurate, fine-grained network representations. </div><div> </div><div>In this dissertation, we discuss a preliminary study and outline three major works that aim to incorporate textual interactions to improve relational representation learning. The preliminary study learns a joint representation that captures the textual similarity in content between interacting nodes. The promising results motivate us to pursue broader research on using social signals for representation learning. The first major component aims to learn explicit node and relation embeddings in social networks. Traditional knowledge graph (KG) completion models learn latent representations of entities and relations by interpreting them as translations operating on the embedding of the entities. However, existing approaches do not consider textual communications between users, which contain valuable information to provide meaning and context for social relationships. We propose a novel approach that incorporates textual interactions between each pair of users to improve representation learning of both users and relationships. The second major component focuses on analyzing how users interact with each other via natural language content. Although the data is interconnected and dependent, previous research has primarily focused on modeling the social network behavior separately from the textual content. In this work, we model the data in a holistic way, taking into account the connections between the social behavior of users and the content generated when they interact, by learning a joint embedding over user characteristics and user language. In the third major component, we consider the task of learning edge representations in social networks. Edge representations are especially beneficial as we need to describe or explain the relationships, activities, and interactions among users. However, previous work in this area lack well-defined edge representations and ignore the relational signals over multiple views of social networks, which typically contain multi-view contexts (due to multiple edge types) that need to be considered when learning the representation. We propose a new methodology that captures asymmetry in multiple views by learning well-defined edge representations and incorporates textual communications to identify multiple sources of social signals that moderate the impact of different views between users.</div>

No description available.

Antenna tilt is the angle subtended by the radiation beam and horizontal plane. This angle plays a vital role in determining the coverage and the interference of the network with neighbouring cells and adjacent base stations. Traditional methods for network optimization rely on rule-based heuristics to do decision making for antenna tilt optimization to achieve desired network characteristics. However, these methods are quite brittle and are incapable of capturing the dynamics of communication traffic. Recent advancements in reinforcement learning have made it a viable solution to overcome this problem but even this learning approach is either limited to its simulation environment or is limited to off-policy offline learning. So far, there has not been any effort to overcome the previously mentioned limitations, so as to make it applicable in the real world. This work proposes a method that consists of transferring reinforcement learning policies from a simulated environment to a real environment i.e. sim-to-real transfer through the use of offline learning. The approach makes use of a simulated environment and a fixed dataset to compensate for the underlined limitations. The proposed sim-to-real transfer technique utilizes a hybrid policy model, which is composed of a portion trained in simulation and a portion trained on the offline real-world data from the cellular networks. This enables to merge samples from the real-world data to the simulated environment consequently modifying the standard reinforcement learning training procedures through knowledge sharing between the two environment’s representations. On the one hand, simulation enables to achieve better generalization performance with respect to conventional offline learning as it complements offline learning with learning through unseen simulated trajectories. On the other hand, the offline learning procedure enables to close the sim-to-real gap by exposing the agent to real-world data samples. Consequently, this transfer learning regime enable us to establish optimal antenna tilt control which in turn results in improved coverage and reduced interference with neighbouring cells in the cellular network. / Antennlutning är den vinkel som dämpas av strålningsstrålen och det horisontella planet. Denna vinkel spelar en viktig roll för att bestämma täckningen och störningen av nätverket med angränsande celler och intilliggande basstationer. Traditionella metoder för nätverksoptimering förlitar sig på regelbaserad heuristik för att göra beslutsfattande för antennlutningsoptimering för att uppnå önskade nätverksegenskaper. Dessa metoder är dock ganska styva och är oförmögna att fånga dynamiken i kommunikationstrafiken. De senaste framstegen inom förstärkningsinlärning har gjort det till en lönsam lösning att lösa detta problem, men även denna inlärningsmetod är antingen begränsad till dess simuleringsmiljö eller är begränsad till off-policy offline inlärning. Hittills har inga ansträngningar gjorts för att övervinna de tidigare nämnda begränsningarna för att göra det tillämpligt i den verkliga världen. Detta arbete föreslår en metod som består i att överföra förstärkningsinlärningspolicyer från en simulerad miljö till en verklig miljö, dvs. sim-till-verklig överföring genom användning av offline-lärande. Metoden använder en simulerad miljö och en fast dataset för att kompensera för de understrukna begränsningarna. Den föreslagna sim-till-verkliga överföringstekniken använder en hybridpolicymodell, som består av en del utbildad i simulering och en del utbildad på offline-verkliga data från mobilnätverk. Detta gör det möjligt att slå samman prover från verklig data till den simulerade miljön och därmed modifiera standardutbildningsförfarandena för förstärkning genom kunskapsdelning mellan de två miljöernas representationer. Å ena sidan möjliggör simulering att uppnå bättre generaliseringsprestanda med avseende på konventionellt offlineinlärning eftersom det kompletterar offlineinlärning med inlärning genom osynliga simulerade banor. Å andra sidan möjliggör offline-inlärningsförfarandet att stänga sim-till-real-klyftan genom att exponera agenten för verkliga dataprov. Följaktligen möjliggör detta överföringsinlärningsregime att upprätta optimal antennlutningskontroll som i sin tur resulterar i förbättrad täckning och minskad störning med angränsande celler i mobilnätet.

<p> Studies to date involving the network properties of drainage systems have been theoretical in nature; and the environmental implications of these network characteristics have not been exploited to the extent that would appear warranted. This situation exists due to the lack of an accurate data source. Many studies have recognized this. inadequacy of the conventional data sources to meet the necessary requirements of efficiency (in data production and handling), accuracy, consistency and uniformity. </p> <p> The present study demonstrates that high altitude, small scale colour infrared photography is capable of providing drainage network data that fulfill all these basic requirements. Data derived from the three drainage basins, mapped from a variety of data sources, demonstrate three important points. The level of detail obtained from the small scale colour infrared photography far exceeds that available from more traditional data sources. Secondly, these network data are statistica+ly consistent with the traditional data sources. Thirdly, the basin characteristics derived from the high altitude data source show a marked association with the known surficial environments and an expected variation from one surficial environment to another. </p> / Thesis / Master of Arts (MA)

High-dimensional data such as multi-modal neuroimaging data and large-scale networks carry excessive amount of information, and can be used to test various scientific hypotheses or discover important patterns in complicated systems. While considerable efforts have been made to analyze high-dimensional data, existing approaches often rely on simple summaries which could miss important information, and many challenges on modeling complex structures in data remain unaddressed. In this proposal, we focus on analyzing structured high-dimensional data, including functional data with important local regions and network data with community structures. The first part of this dissertation concerns the detection of ``important'' regions in functional data. We propose a novel Bayesian approach that enables region selection in the functional data regression framework. The selection of regions is achieved through encouraging sparse estimation of the regression coefficient, where nonzero regions correspond to regions that are selected. To achieve sparse estimation, we adopt compactly supported and potentially over-complete basis to capture local features of the regression coefficient function, and assume a spike-slab prior to the coefficients of the bases functions. To encourage continuous shrinkage of nearby regions, we assume an Ising hyper-prior which takes into account the neighboring structure of the bases functions. This neighboring structure is represented by an undirected graph. We perform posterior sampling through Markov chain Monte Carlo algorithms. The practical performance of the proposed approach is demonstrated through simulations as well as near-infrared and sonar data. The second part of this dissertation focuses on constructing diversified portfolios using stock return data in the Center for Research in Security Prices (CRSP) database maintained by the University of Chicago. Diversification is a risk management strategy that involves mixing a variety of financial assets in a portfolio. This strategy helps reduce the overall risk of the investment and improve performance of the portfolio. To construct portfolios that effectively diversify risks, we first construct a co-movement network using the correlations between stock returns over a training time period. Correlation characterizes the synchrony among stock returns thus helps us understand whether two or multiple stocks have common risk attributes. Based on the co-movement network, we apply multiple network community detection algorithms to detect groups of stocks with common co-movement patterns. Stocks within the same community tend to be highly correlated, while stocks across different communities tend to be less correlated. A portfolio is then constructed by selecting stocks from different communities. The average return of the constructed portfolio over a testing time period is finally compared with the SandP 500 market index. Our constructed portfolios demonstrate outstanding performance during a non-crisis period (2004-2006) and good performance during a financial crisis period (2008-2010). / PHD / High dimensional data, which are composed by data points with a tremendous number of features (a.k.a. attributes, independent variables, explanatory variables), brings challenges to statistical analysis due to their “high-dimensionality” and complicated structure. In this dissertation work, I consider two types of high-dimension data. The first type is functional data in which each observation is a function. The second type is network data whose internal structure can be described as a network. I aim to detect “important” regions in functional data by using a novel statistical model, and I treat stock market data as network data to construct quality portfolios efficiently

The goal of this thesis is to analyze the current state of the customer's computer network and his requirements. Then the changes and the way of their realization were proposed. The next part of the thesis are the description of appropriate components, selected technologies and the detailed description of the implementation process, including blueprints.

The goal of this thesis is to introduced the principles of access control technologies, the General Data Protection Regulation and the software for data leakage protection. An essential part of the work is a draft and implementation of the expansion module for user device authentication including shared storage access authorization. Therefore, this module allows to verify whether a user can access shared corporate resources. It also allows to enable or disable access based on specified attributes, such as the type of the protected service or user permission. The basic verification of the module's functionality is realized through different sets of tests and a virtual environment that simulates the corporate environment. The result of the draft is a module that allows to verify access based on the device, and this module is moreover integrated into the Safetica security platform.

We consider a wireless sensor network, in which end users are interested in maximizing the useful information supplied by the network till network partition due to inevitable node deaths. Neither throughput maximization nor network lifetime maximization achieves the objective: A network with high throughput provides information at a high rate, but can exhaust the nodes of their energies quickly; similarly, a network can achieve a long lifetime by remaining idle for most of the time. We propose and seek to maximize a new metric: “Aggregate bit transported before network partition” (a product of throughput and lifetime), which precisely captures the usefulness of sensor networks. We model the links in the wireless sensor network as wired links with reduced equivalent capacities, formulate and solve the problem of maximizing bits transported before network partition on arbitrary networks. To assess the benefits that network coding can yield for the same objective, we study a scenario where the coding-capable nodes are placed on a regular grid. We propose an optimal algorithm to choose the minimum number of coding points in the grid to ensure energy efficiency. Our results show that, even with simple XOR coding, the bits transported can increase up to 83 % of that without coding. Further, we study the problem of in-network data aggregation in a wireless sensor network to achieve minimum delay. The nodes in the network compute and forward data as per a query graph, which allows operations belonging to a general class of functions. We aim to extract the best sub-network that achieves the minimum delay. We design an algorithm to schedule the sub-network such that the computed data reaches sink at the earliest. We consider directed acyclic query graphs as opposed to the existing work which considers tree query graphs only.