The anti-forensics process using different methods to hide information or alter it which can lead theinvestigation to the wrong direction. The rising of new skills and experience levels can put the wholeinvestigation in jeopardy especially in the aspect of financial and time loss. Anti-forensics toolsleveraging the cryptographic techniques, data deletion and more novel methods to counter theforensics tools. In this thesis we will examine the techniques that are used by criminals to break downthe cybercrime investigation and all related digital forensics. The focus in this theses will be theeffectiveness of the anti-forensics techniques in hiding traces from the forensics tools.

No description available.

The expansion of uniquely identifiable devices that can report their environment through the internet has beenrapid. The Internet of Things(IoT) is the next step in interconnectivity between humans and devices. Toaccommodate the huge data flow and big data processing that follows with IoT cloud computing serves as theinfrastructure and as the platform to handle and process the data produced by IoT devices. These cloud basedIoT platforms can be commercial, as is the case with Google Cloud Platform, Microsoft Azure or Amazon AWS,or open source like FIWARE. Two solutions have been created and tested for an IoT environment using boththe commercial alternative AWS and the open source alternative FIWARE. The solutions showed great resultswhen performing the tests. When creating and developing these IoT solutions AWS offers a better developerexperience than that of FIWARE. This is because of the intuitive user interface as well as the integration anddocumentation of their own components. This thesis shows how to connect information gathering sensors totwo(2) different systems, one which uses the FIWARE IoT platform and one that uses the AWS platform withAWS IoT Core, AWS Rules engine and AWS DynamoDB. This thesis also evaluates the two systems based onqualitative and quantitative analysis. The conclusion of this thesis is that the two systems have differences andcommonalities and that the AWS system is the preferable one based on the expansive documentation, ease ofuse and future support of the AWS platform. Whilst AWS proves to be the preferable platform FIWARE lookslike a suitable open source alternative if the documentation of the FIWARE components get updated in thenear future.

The cloud has become much more important and relevant to the IT industry in recentyears. Instead of buying and maintaining their own physical servers, companies now often opt for renting servers and services from cloud providers. These servers can be thought ofas abstract units of computing power that can be dynamically allocated or disposed of de-pending on need. Configuring such an infrastructure can be very complex and challenging.This is where Infrastructure as Code (IaC) comes into play. Instead of having to manually create and configure each resource and their web of con-nections, IaC can be used to describe the desired state of the infrastructure in a declarativeway as code. The IaC tools will then take care of creating and configuring the resources inthe cloud to match that desired state. There exists quite a few different IaC tools. This thesis will go through the IaC toolsAWS CDK and Pulumi and compare them to provide you with insight to help you choosewhich tool will fit you best. The focus will lie on the readability aspect of the tools. This comparison was done by implementing a specific infrastructure using each ofthe tools and then comparing the implementations using a survey where people voted onwhich one they preferred as well as by using the metrics: Lines of Code (LOC), CyclomaticComplexity (CC) and Cognitive Complexity (COG). The survey results showed that AWS CDK was generally preferred in terms of read-ability due to higher level abstractions and nicely provided resource defaults. However,Pulumi’s way of configuring stack specific variables using dedicated YAML files was pre-ferred as well as declaring the resources at the level, compared to inside a class as in AWSCDK. AWS CDK also had a better metrics score when looking at LOC and Cognitive Com-plexity, where Pulumi had a better CC score. The command line interface (CLI) of the tools were also briefly evaluated to see if theyfollowed some of the recommended guidelines. Both AWS CDK and Pulumi followed mostof the guidelines. However, neither used stdout and stderr correctly and Pulumi did nothave an explicit flag for extracting the current CLI version.

Social media platforms have during the last decade expanded immensely in both the number of users and posts. An extensive number of previous studies have observed how users engage on these platforms to study how social systems work. However, compared to other social media platforms, Instagram has been less extensively researched. This work analyses user engagement on the social media platform Instagram for different characteristics related to posts, uploaders, and media files, consisting of photos and videos. What differentiates this work from others is studying the temporal dynamics of users' engagement across albums, photos, and videos. The results show that album posts receive the highest number of interactions and have the longest engagement lifespan, which is followed by photo posts. Additionally, the most important characteristic that attracts users' interactions is related to the uploader and includes their social network size and uploading rate. Further, different categories of users are analysed with respect to the post type. Compared to other influencer groups, brands and other types of organisations receive fewer interactions and musicians tend to have more loyal followers on Instagram. The conducted analysis may influence brands' and influencers' marketing strategies on various social media platforms and the result can influence the creation of analytical models to predict the temporal dynamics of user engagement on Instagram.

No description available.

Design by contract, abbreviated as DBC, is a software development methodology that aims to increase the reliability and robustness of software components. While a fair amount of research has been done around how DbC can be utilised in an in-process object-oriented system, not much is known about how DbC concepts can be applied to systems that predominantly communicate over the network by way of remote procedure calls. With the recent increase in popularity of service-oriented and microservice system architectures, the ability to develop robust networked components at scale is highly relevant. This study applies the DbC notion of software contracts to remote procedure calls by developing an interface definition language that can be used in conjunction with JSON-RPC and JSON Schema. The results demonstrate that it is possible to leverage DbC concepts when implementing networked software services, but that it may in many cases be impractical to do so due to the resulting concurrency issues and increased complexity. / Design genom kontrakt, abbrevierat ned till DbC, är en metodik för mjukvaruutveckling vars syfte är att öka pålitlighet och robusthet av programvara. Medan en ansenlig mängd har forskats för att bedöma hur DbC kan utnyttjas i ett objekt-orentierad sammanhang, det är fortsatt ovetandes om hur DbC konceptet kan appliceras till system som huvudsakligen kommunicerar över nätverket. Med den ökade populariteten av service-orienterad mjukvaruarkitektur, förmågan att utveckla robusta nätverkskomponenter är högst relevant. Denna studie applicerar en DbC förståelse av mjukvarukontrakt till gränssnittav av avlägsna proceduranropp genom att använda JSON-RPC och JSON Schema. Resultatet visar att det finns möjlighet att verkställa DbC koncepter när man implementerar mjukvara som kommunicerar över nätverket, men detta kan vara opraktiskt pga ökad komplexitet och resulterande samtidighetsproblem.

A common task within research of neuronal morphology is neuron touch detection, that is finding the points in space where two neurites approach each other to form a synapse. In order to make efficient use of cache memory, it is important to store points that are close in space close in memory. One data structure that aims to tackle this complication is the R*-tree. In this thesis, a spatial query for touch detection was implemented and the scalability of the R*-tree was estimated on realistic neuron densities and extrapolated to explore execution times on larger volumes. It was found that touch detection on this data structure scaled much like the optimal algorithm in 3D-space and more specifically that the computing power needed to analyze a meaningful portion of the human cortex is not readily available. / En vanlig uppgift inom forskning av neuronal morfologi är att hitta var mellan olika neuroner synapser bildas. För att använda cache-minne effektivt är det viktigt att lagra punkter som är nära i rymden nära i minnet. En datastruktur som ämnar att lösa detta är R*-trädet. I denna rapport så implementerades en sökning av rummet och skalbarheten för R*-trädet uppskattades på realistiska neurondensiteter för att sedan extrapoleras och utforska körtider på större volymer. Det konstaterades att denna datastruktur skalade sig mycket som den optimala algoritmen i tredimensionell rymd och mer specifikt att datorkraften som behövs för att analysera en meningsfull del av människans hjärnbark inte är fritt tillgänglig.

The brain is the most complex organ in the human body, and there are many reasons to study it. One way of studying the brain is through digital simulations. Such simulations typically require large amounts of data on the 3dimensional structure of individual neurons to be stored and processed efficiently. Commonly, such data is stored using data structures for spatial indexing such as R-trees. A typical operation within neuroscience which needs to make use of these data structures is the range query: a search for all elements within a given subvolume of the model. Since these queries are common, it is important they can be made efficiently. The purpose of this study is to compare a selection of construction methods (repeated R*-tree insertion, one-dimensional sorting packing, Sort-Tile-Recursive (STR) packing, Adaptive STR packing, Hilbert/Z-order/Peano curve packing, and binary splitting packing) for R-trees with respect to their performance in terms of building time, query page reads and query execution time. With reconstructions of neurons from the human brain, ten datasets were generated with densities ranging from 5,000 to 50,000 neurons/mm3 in a 300 µm 600 µm 300 µm volume. Range queries were then run on the R-trees constructed from these datasets. The results show that the lowest query times were achieved using STR packing and Adaptive STR packing. The best performing construction techniques in terms of build time were Peano and Z-order curve packing. / Hjärnan är kroppens mest komplicerade organ och det finns många anledningar att studera den. Ett sätt att studera hjärnan är genom datorsimuleringar. Sådana datorsimuleringar kräver en stor mängd tredimensionell neurondata som behöver lagras och behandlas effektivt. R-träd är en vanlig datastruktur för att behandla sådan data, och en vanlig operation man inom neurovetenskapen vill genomföra är sökningar efter element i en given delvolym av modellen man arbetar med. Det är därför viktigt att dessa operationer kan genomföras effektivt. Syftet med denna studie är att jämföra ett urval av konstruktionstekniker (upprepad R*-trädsinsättning, endimensionell sorteringspackning, STR-packning, adaptiv STR-packning, Hilbert-packning, Zordningspackning, Peano-kurvpackning samt binär klyvpackning) för R-träd med avseende på prestanda i konstruktionstid, antal sidinläsningar och exekveringstid för sökningar. Tio datamängder genererades med nervcellsdensiteter mellan 5,000 och 50,000 celler/mm3 i en volym på 300 µm 600 µm 300 µm. Dessa användes sedan för konstruktion av olika R-träd i vilka en sekvens av delvolymssökningar gjordes. Resultaten visar att den lägsta söktiden erhölls med R-träd konstruerade genom STR-packning och adaptiv STR-packning, medan konstruktionstiden var som lägst för packning med Peanokurvan och Z-ordningskurvan.

An ongoing problem regarding the automatic classification of neurons by their morphology is the lack of consensus between experts on neuron types. Unsupervised clustering using persistent homology as a descriptor for the morphology of neurons helps tackle the problem of bias in feature selection and has the potential of aiding neuroscience research in developing a framework for automatic neuron classification. This thesis investigates how two different unsupervised machine learning algorithms would cluster persistence images of already labeled neurons and how similar their clusterings would be. The results showed that the clusterings done by both methods were highly similar and that there was a large variation within the neuronal types defined by experts. / Ett pågående problem gällande den automatiska klassificeringen av neuroner med avseende på morfologi är bristen på konsensus bland experter vad gäller neurontyper. Oövervakad klusteranalys med persistent homologi som en deskriptor för neuroners morfologi hjälper lösa problemet med partiskhet inom egenskapsurval och kan potentiellt gynna neurovetenskapen i utvecklingen av ett ramverk för automatisk klassificering av neuroner. Denna uppsats hade som mål att undersöka hur två olika oövervakade maskininlärningsalgoritmer klassificerar persistensbilder av tidigare klassificerade neuroner samt graden av överensstämmelse mellan de två metoderna. Studiens resultat visade att båda metoders resultat hade en hög grad av överensstämmelse och visade även på en stor variation inom de klasser av neuroner som redan definierats av experter.