| 2020-06-01 - ADIT |
QUIC Behavior on Dual Connectivity
David Hasselquist, Christoffer Lindström
Avancerad (30hp)
kl 13:15, https://liu-se.zoom.us/j/69013602242?pwd=R1I3aWE1cWRwWXpFa2Z3ZFZUQ3JQQT09 (In English)
[Abstract]QUIC is a relatively new transport layer network protocol that has gained popularity over the last few years. The protocol was initially developed by Google and standardization work has been continued by the Internet Engineering Task Force (IETF) with the goal of it becoming the next generation transport protocol. While the standardization work is not yet finished, the protocol has seen a large adoption, already covering a large portion of the internet traffic. As a new protocol, many researchers have studied QUIC and compared it to TCP in typical scenarios. However, few studies have been performed on QUIC in specific scenarios. In this thesis, we present the first performance study of QUIC over Dual Connectivity (DC). DC is a multi-connectivity technique that allows users to connect to multiple cell towers with one user equipment. It is an important lower-layer feature accelerating the transition from 4G to 5G, which is also expected to play an important role in standalone 5G networks. With DC, higher throughput and reliability can be achieved by using multiple paths simultaneously. However, the drawback of DC is that it introduces packet reordering and jitter, which can significantly impact the performance of upper-layer protocols such as TCP and QUIC.
To study the extent of this effect, a testbed is set up to evaluate QUIC over DC. Our performance evaluation compares the throughput of QUIC over DC with that of TCP over DC, and evaluates the fairness of QUIC over DC. Using a series of throughput and fairness experiments, we show how QUIC is affected by different DC parameters, network conditions, and whether the DC implementation aims to improve throughput or reliability. Our findings provide network operators with insights into understanding the impacts of splitting QUIC traffic in a DC environment. We show the value of increasing the UDP receive buffers when running QUIC over DC and that QUIC can utilize the increased bandwidth and reliability in DC, provided that the links' characteristics are similar. We also show that with reasonably selected DC parameters and increased UDP receive buffers, QUIC over DC performs similarly to TCP over DC and achieves optimal systemwide fairness under symmetric link conditions when DC is not used for packet duplication. |
| 2020-06-03 - ADIT |
Ett skyddsnät för onlinekasino - Att predicera risken för spelproblem med hjälp av Bayesianska nätverk
Kristian Sikiric´
Avancerad (30hp)
kl 13:15, https://liu-se.zoom.us/j/69435420466?pwd=M3YrQmZCRlhSY0hCeGloWFlWUkJvZz09 (På svenska)
[Abstract]As online casino and betting increases in popularity across the globe, the importance of green gambling has become an important subject of discussion. The Swedish betting company, ATG, realises the benefits of this and would like to prevent their gamblers from falling into problem gambling. To predict problem gambling, Bayesian networks were trained on previously identified problem gamblers, separated into seven risk groups. The network was then able to predict the risk group of previously unseen gamblers with an accuracy of 94%. It also achieved an average precision of 89%, an average recall of 96% and an average f1-score of 93%. The features in the data set were also ranked, to find which were most important in predicting problem gambling. It was found that municipality, which day of the week the transaction was made and during which hour of the day were the most important features. Also, the Bayesian network was also made as simple as possible, by removing irrelevant features and features which carry very low importance. |
| 2020-06-04 - STIMA |
Probabilistic Regression using Conditional Generative Adversarial Networks
Joel Oskarsson
Avancerad (30hp)
kl 10:15, https://liu-se.zoom.us/j/64533079221?pwd=Q05qZ0hyaTAzK0daZXRGc3FHQlpoUT09 (In English)
[Abstract]Regression is a central problem in statistics and machine learning with applications everywhere in science and technology. In probabilistic regression the relationship between a set of features and a real-valued target variable is modelled as a conditional probability distribution. There are cases where this distribution is very complex and not properly captured by simple approximations, such as assuming a normal distribution. This thesis investigates how conditional Generative Adversarial Networks (GANs) can be used to properly capture more complex conditional distributions. GANs have seen great success in generating complex high-dimensional data, but less work has been done on their use for regression problems. This thesis presents experiments to better understand how conditional GANs can be used in probabilistic regression. Different versions of GANs are extended to the conditional case and evaluated on synthetic and real datasets. |
| 2020-06-04 - SaS |
Net benefits analysis of Visual Regression Testing in a Continuous Integration environment: An industrial case study
Axel Löjquist
Avancerad (30hp)
kl 15:00, https://liu-se.zoom.us/j/61645977078?pwd=d0EvV2hxUkN3Y2daTjA3eHk0TDk3QT09 (In English)
[Abstract]Maintaining quality in software is a difficult task for several reasons, such as, company growth, time-to-market demands, code complexity and more. GUI testing tools and Continuous Integration (CI) are common practice today to tackle some of the issues with maintaining software quality. However, these techniques bring a set of challenges. Visual Regression Testing (VRT) is a special kind of GUI testing technique focused upon image-based assertions. This study presents an implementation and investigation of the benefits and drawbacks of introducing VRT for a CI environment in an industrial context. Additionally, the thesis investigates factors that need to be considered upon this transition. The results show that benefits are associated with this approach, such as, quicker feedback times and an increase in testing frequency. However, drawbacks and implications were also identified, such as, test maintenance and organizational concerns, indicating careful consideration needs to be taken by an organization before proceeding with an implementation. |
| 2020-06-04 - HCS |
Voice Assistants and Railway Passengers - A user-centered exploration of value creation opportunities in a railway service context
Gustav Backhans, Douglas Driving
Avancerad (30hp)
kl 15:00, https://liu-se.zoom.us/j/68188100756?pwd=UmJVZGxnZXQveXlsbVNXNUV3NEtlZz09 (På svenska)
[Abstract]There is a current rising trend of using voice assistants (VAs) to perform tasks in new ways, and various companies are considering introducing them as part of their service propositions. One such company is the railway service provider SJ AB that is interested in knowing and understanding how a VA may benefit their passengers. In order to better understand the utility of a VA in the railway service context, this thesis aims to explore the value creation opportunities that the introduction of such a technology presents. This exploration is done through identifying the potential functionality of such a VA, what value that functionality can create for passengers, and what barriers exist for creating that value. This identification is done through a user-centered research process, during which design probes and experience prototyping were carried out with railway passengers and analysed through qualitative content analysis.
The results show that a VA presents fast, convenient and intuitive access to a wide set of functions, but is hindered by its invisible affordances and the user’s preconceptions. It shares several functions with existing channels, presented in a new way, that span the entire course of the train journey. The functionality and the identified functions primarily create utilitarian values, which connects to the passengers pragmatic view of the service. Furthermore, the passengers perceived the VA as an agent able to take over responsibilities and tasks from the passenger, resulting in a peace of mind but also a diminished sense of control. Finally, the railway service context affected what functionality is suitable, what values can be created and what barriers need to be considered. Some of the value creation possibilities and barriers are also contemporary and might change with shifting social norms and further technological development. |
| 2020-06-05 - STIMA |
Evaluating Hierarchical LDA Topic Models for Article Categorization
Jennifer Lindgren
Avancerad (30hp)
kl 08:15, https://liu-se.zoom.us/j/66565611940 (In English)
|
| 2020-06-05 - SaS |
Analys av Microservice-arkitekturer med avseende på ...
Jesper Holmström
Avancerad (30hp)
kl 08:30, https://liu-se.zoom.us/j/63661977611?pwd=bXJPQnVEV1JHYVM3UnByMmtpbHNFQT09 (In English)
[Abstract]The microservice architecture is a new architectural style that structures an application into a set of small, independently deployable microservices, as opposed to the traditional monolith approach with a single executable. The microservice architecture is a distributed system that results in new challenges and increased complexity. This study expands the previous related research and investigates the implications of using the one-database-per-service pattern and a solution to the introduced need of queries spanning multiple microservices. In this thesis, two applications are presented, one with the microservice architecture and one monolithic counterpart, which are compared in terms of response time and throughput. As a solution for the distributed queries, the API composition pattern was chosen. The results of the experiments conclude in a greater understanding of the difficulty in distributed queries as well as the benefits and limitations of the API composition pattern. It shows that the API composition pattern is a valid solution for distributed queries. However, it does perform worse in terms of response time and throughput than the monolith prototype. This results in the insights that one must carefully choose, with respect to the requirements of the system, when to apply it. |
| 2020-06-05 - SaS |
Architectural Implications of Serverless and Function-as-a-Service
Oscar Andell
Avancerad (30hp)
kl 10:00, https://liu-se.zoom.us/j/68297618738?pwd=VVNLc1NDbk5heko1OFl3cFJ3blVRQT09 (In English)
[Abstract]Serverless or Function-as-a-Service (FaaS) is a recent architectural style that is based on the principles of abstracting infrastructure management and scaling to zero, meaning application instances are dynamically started and shut down to accommodate load. This concept of no idling servers and inherent autoscaling comes with benefits but also drawbacks.
This study presents an evaluation of the performance and implications of the serverless architecture and contrasts it with the so-called monolith architectures. Three distinct architectures are implemented and deployed on the FaaS platform Microsoft Azure Functions as well as the PaaS platform Azure Web App. Results were produced through experiments measuring cold starts, response times, and scaling for the tested architectures as well as observations of traits such as cost and vendor lock-in. The results indicate that serverless architectures, while it is subjected to drawbacks such as vendor lock-in and cold starts, provides several benefits to a system such as reliability and cost reduction. |
| 2020-06-05 - HCS |
A Comparison between Different Recommender System Approaches for a Book and Author Recommender System
Jesper Hedlund, Emma Nilsson Tengstrand
Avancerad (30hp)
kl 15:00, https://liu-se.zoom.us/j/61378084983?pwd=czlJODZwR0p4UkRwZ1FsMkl2U3hHZz09 (På svenska)
[Abstract]A recommender system is a popular tool used by companies to increase customer satisfaction and to increase revenue. Collaborative filtering and content-based filtering are the two most common approaches when implementing a recommender system, where the former provides recommendations based on user behaviour, and the latter uses the characteristics of the items that are recommended.
The aim of the study was to develop and compare different recommender system approaches, for both book and author recommendations and their ability to predict user ratings of an e-book application. The evaluation of the models was done by measuring RMSE and MAE. Two pure models were developed, one based on collaborative filtering and one based on content-based filtering. Also, three different hybrid models using a combination of the two pure approaches were developed and compared to the pure models. The study also explored how an aggregation of book data to author level could be used to implement an author recommender system.
The results showed that the aggregated author data was more difficult to predict. However, it was difficult to draw any conclusions of the performance on author data due to the data aggregation. It was though clear that it was possible to derive author recommendations based on data from books. The study also showed that the collaborative filtering model performed better than the content-based filtering model according to RMSE but not according to MAE. The lowest RMSE and MAE, however, were achieved by combining the two approaches in a hybrid model. |
| 2020-06-08 - HCS |
Predicting the Financial Impact of the CEO’s Comments in Quarterly Reports
Ludvig Westerdahl
Avancerad (30hp)
kl 10:00, https://liu-se.zoom.us/j/61757150962?pwd=SG5Rb1VNTGJGTllETlNxekE5cUY3UT09 (På svenska)
[Abstract]This thesis investigates how the CEO’s comments in quarterly reports affect the finan- cial performance of a company by predicting their stock price using machine learning and natural language processing. The dataset used consists of historical information such as the stock price (quantitative) and the CEO’s comments (qualitative). Where the qualitative information was embedded using the paragraph vector document embedding technique and used with the quantitative data in three type of models. The models tested was Sup- port Vector Machine and Artificial Neural Network against a Naive Bayes baseline. Fur- ther, each model was trained and evaluated using the quantitative, qualitative and both datasets and the results were confirmed using statistical significant testing. Finally, the best models from the evaluation step were used to simulate a trading strategy to buy the stock if the model predicted that the price of the stock would rise. The statistically signifi- cant improvements of using the CEO’s comments and the hypothetical profits the trading strategies rendered show that the CEO’s comments adds some predictive ability in terms of their stock price and thus their financial performance. |
| 2020-06-09 - HCS |
Autonomous cleaning robot as a service – Mapping current cleaning service and prototyping future cleaning service that uses autonomous robotics
Jeff Pertot, Pernilla Sandén
Avancerad (30hp)
kl 13:15, https://liu-se.zoom.us/j/68788403024?pwd=Vk14bnFzZ0QrZGtNS0V5K25LNkxPQT09 (In English)
|
| 2020-06-10 - HCS |
Automated Image Suggestions for News Articles - An evaluation of text and image representations in an image retrieval system
Pontus Svensson
Avancerad (30hp)
kl 10:00, https://liu-se.zoom.us/j/68411179812?pwd=blo3UlpIbndtb2RIM0FqS3g3cnJidz09 (På svenska)
[Abstract]Multimodal machine learning is a subfield of machine learning that aims to relate data from different modalities, such as texts and images. One of many applications that could be built upon this technique is an image retrieval system that, given a text query, retrieves suitable images from a database. In this thesis, a retrieval system based on canonical correlation is used to suggest images for news articles. Different dense text representations produced by Word2vec and Doc2vec, and image representations produced by pre-trained convolutional neural networks are explored to find out how they affect the suggestions. Which part of an article is best suited as query to the system is also studied. In addition, experiments are carried out to determine if an article's date of publication can be used to improve the suggestions. The results show that Word2vec outperforms Doc2vec in the task, which indicates that the meaning of article texts are not as important as the individual words they consist of. Furthermore, the queries are improved by rewarding words that are particularly significant. |
| 2020-06-10 - SaS |
Migration to P4-Programmable Switches and Implementation of the Rapid Spanning Tree Protocol
Henrik Lindström
Avancerad (30hp)
kl 10:15, https://liu-se.zoom.us/j/61023746240 (In English)
[Abstract]P4 is a high-level language for programming the forwarding planes of network switches. These P4-programmable switches come with no pre-defined behavior or protocols, so it is entirely up to the loaded P4 program to define these. This allows the user to exclude any unwanted functionality and to create custom protocols. It also removes the dependence on the switch vendor in terms of both trust and addition of new features.
This thesis looks at migration from traditional switches to P4-programmable ones. Since no behavior is included out-of-the-box in the P4 switches, a search is made for open-source P4 projects and the functionality they provide is evaluated. It is found that most link layer functionality can be achieved with them, with the exception being loop prevention by spanning tree protocols. Therefore, one of the projects is extended with an implementation of the Rapid Spanning Tree Protocol based on the IEEE 802.1D-2004 standard. Finally, partial migration of networks to P4 switches and to the Software Defined Networking (SDN) paradigm is studied based on a literature review. Four general approaches and specific implementations of these are found, and it is concluded that such a hybrid network can still benefit from P4 and having a centralized SDN controller. |
| 2020-06-10 - HCS |
Exploring Emerging Entities and Named Entity Disambiguation in News Articles
Robin Ellgren
Avancerad (30hp)
kl 13:00, https://liu-se.zoom.us/j/64829873933?pwd=UHVmOWx4SzJqOVVNTzduZXd2SlAxdz09 (På svenska)
[Abstract]Publicly editable knowledge bases such as Wikipedia and Wikidata have over the years grown tremendously in size. Despite the quick growth, they can never be fully complete due to the continuous stream of events happening in the world. In the task of Entity Linking, it is attempted to link mentions of objects in a document to its respective corresponding entry in a knowledge base. However, due to the incompleteness of knowledge bases, new or emerging entities cannot be linked. Attempts to solve this issue have created the field referred to as Emerging Entities. Recent state-of-the-art work have addressed the issue with promising results in English. In this thesis, the previous work is examined by evaluating its method in the context of a much smaller language; Swedish. The results reveal an expected drop in overall performance although remaining relative competitiveness. This indicates that the method is a feasible approach to the problem of Emerging Entities even for much less used languages. Due to limitations in the scope of the related work, this thesis also suggests a method for evaluating the accuracy of how the Emerging Entities are modeled in a knowledge base. The study also provides a comprehensive look into the landscape of Emerging Entities and suggests further improvements. |
| 2020-06-10 - HCS |
Service Design of a Communication Tool for Overcoming Language Barriers in Caregiver-Patient Consultations
Ida Samuelsson, Emma Svensson
Avancerad (30hp)
kl 13:15, https://liu-se.zoom.us/j/65044340833?pwd=YnV3QWU2Snhkb2NTZERGdk1wSTd4UT09 (På svenska)
[Abstract]The purpose of this study is to explore the design of a digital communication tool for patient-caregiver communication with language barriers. This is done through a design case at Worldish, a company that develops a digital communication tool named Helen. An iterative service design approach is used in the case study to understand enablers and complicators for usage of a digital communication tool in care consultations with language barriers, as well as how the design of such tool can better be adapted to fit the service process of a care consultation.
The case study is divided into phases where the first two, phase A and B, investigates the care consultation process with and without language barriers as well as how a digital communication tool is used in this process. These phases results in several insights about what in the care consultation process with language barriers that enable and complicates the usage of a digital communication tool. In total, this study identifies four enabling factors and six complicating factors for usage of a digital communication tool in the care consultation process.
In the third and fourth phase, phase C and D, two design-iterations are conducted where design ideas are ideated, prototyped and tested with users. These phases results in design features and insights connected to how the design of a digital communication tool can better be adapted to the service process. In total this study identifies six guidelines that serves as inspiration for someone who wants to create or adapt the design of a digital communication tool to the service process in a care consultation. |
| 2020-06-10 - SaS |
Visualization of machine learning data for radio networks. A case study at Ericsson.
Bingyu Niu
Grundnivå (16hp)
kl 13:15, https://teams.microsoft.com/l/meetup-join/19%3ameeting_YTY0MDdhNmEtNTY3MC00ZWM1LTk0MDAtY2JlNzEyMGI3MzU0%40thread.v2/0?context=%7b%22Tid%22%3a%22913f18ec-7f26-4c5f-a816-784fe9a58edd%22%2c%22Oid%22%3a%22a4398856-ea65-4223-96e2-2d96d76ce828%22%7d (In English)
[Abstract]This thesis presents an approach to develop a visualization software for time-varying and geographic-based data. The machine learning team at Ericsson has collected data from their machine learning algorithms. The data set contains time and geographic information. To have a better understanding of the result made by the machine learning algorithms, it is important to understand the pattern of the data. It is hard to see the pattern of the data by only looking at the raw data set, and data visualization software will help the users to have a more understandable view of the data. To choose a suitable GUI library, three common GUI libraries were compared. The Qt framework was chosen as the GUI library and development framework because of its wide-range support to user interface design. Animation is the main method to visualize the data set. The performance evaluation of the software shows that it handles the backend data efficiently, renders fast in the frontend and has low memory and CPU usage. The usability testing indicates that the software is easy to use. In the end, the thesis compares its method to a previous method, developed in R. The comparison shows that even though the old method is easier to develop, it has worse performance. |
| 2020-06-10 - HCS |
Identifying Content Blocks on Web Pages using Recursive Neural Networks and DOM-tree Features
Teodor Riddarhage
Avancerad (30hp)
kl 15:00, https://liu-se.zoom.us/j/67944195474?pwd=SFJMdEVNMTNIdXhuRFlxQ0VIYjk5QT09 (In English)
[Abstract]The internet is a source of abundant information spread across different web pages. The identification and extraction of information from the internet has long been an active area of research for multiple purposes relating to both research and business intelligence. However, many of the existing systems and techniques rely on assumptions that limit their general applicability and negatively affect their performance as the web changes and evolves. This work explores the use of Recursive Neural Networks (RecNNs) along with the extensive amount of features present in the DOM-trees for web pages as a technique for identifying information on the internet without the need for strict assumptions on the structure or content of web pages. Furthermore, the use of Sparse Group LASSO (SGL) is explored as an effective tool for performing feature selection in the context of web information extraction. The results show that a RecNN model outperforms a similarly structured feedforward baseline for the task of identifying cookie consent dialogs across various web pages. Furthermore, the results suggest that SGL can be used as an effective tool for feature selection of DOM-tree features. |
| 2020-06-11 - HCS |
News Value Modeling and Prediction using Textual Features and Machine Learning
Rebecca Lindblom
Avancerad (30hp)
kl 13:00, https://liu-se.zoom.us/j/67794693797?pwd=VXM5Vk5ITytrQkpyczNMTEZXQU1DQT09 (På svenska)
[Abstract]News value assessment has been done forever in the news media industry and is today often done in real-time without any documentation. Editors take a lot of different qualitative aspects into consideration when deciding what news stories will make it to the first page. This thesis explores how the complex news value assessment process can be translated into a quantitative model, and also how those news values can be predicted in an effective way using machine learning and NLP.
Two models for news value were constructed, for which the correlation between modeled and manual news values was measured, and the results show that the more complex model gives a higher correlation. For prediction, different types of features are extracted, Random Forest and SVM are used, and the predictions are evaluated with accuracy, F1-score, RMSE, and MAE. Random Forest shows the best results for all metrics on all datasets, the best result being on the largest dataset, probably due to the smaller datasets having a less even distribution between classes. |
| 2020-06-17 - HCS |
Realtime Web Audio Streaming
Dennis Bennhage, William Utbult
Grundnivå (16hp)
kl 13:15, Distans (På svenska)
|
| 2020-06-17 - HCS |
Cross-platform development: A performance comparison between React Native and Cordova
Gustav Leffler
Grundnivå (16hp)
kl 13:15, Distans (På svenska)
|
| 2020-06-17 - HCS |
A concept for quality control by using microservice
Mats Eriksson
Grundnivå (16hp)
kl 13:15, Distans (På svenska)
|
Swedish web site
