Filtern
Dokumenttyp
- Masterarbeit (19) (entfernen)
Schlagworte
Institut
- Institute for Web Science and Technologies (19) (entfernen)
This thesis explores and examines the effectiveness and efficacy of traditional machine learning (ML), advanced neural networks (NN) and state-of-the-art deep learning (DL) models for identifying mental distress indicators from the social media discourses based on Reddit and Twitter as they are immensely used by teenagers. Different NLP vectorization techniques like TF-IDF, Word2Vec, GloVe, and BERT embeddings are employed with ML models such as Decision Tree (DT), Random Forest (RF), Logistic Regression (LR) and Support Vector Machine (SVM) followed by NN models such as Convolutional Neural Network (CNN), Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) to methodically analyse their impact as feature representation of models. DL models such as BERT, DistilBERT, MentalRoBERTa and MentalBERT are end-to-end fine tuned for classification task. This thesis also compares different text preprocessing techniques such as tokenization, stopword removal and lemmatization to assess their impact on model performance. Systematic experiments with different configuration of vectorization and preprocessing techniques in accordance with different model types and categories have been implemented to find the most effective configurations and to gauge the strengths, limitations, and capability to detect and interpret the mental distress indicators from the text. The results analysis reveals that MentalBERT DL model significantly outperformed all other model types and categories due to its specific pretraining on mental data as well as rigorous end-to-end fine tuning gave it an edge for detecting nuanced linguistic mental distress indicators from the complex contextual textual corpus. This insights from the results acknowledges the ML and NLP technologies high potential for developing complex AI systems for its intervention in the domain of mental health analysis. This thesis lays the foundation and directs the future work demonstrating the need for collaborative approach of different domain experts as well as to explore next generational large language models to develop robust and clinically approved mental health AI systems.
This thesis focuses on approximate inference in assumption-based argumentation frameworks. Argumentation provides a significant idea in the computerization of theoretical and practical reasoning in AI. And it has a close connection with AI, engaging in arguments to perform scientific reasoning. The fundamental approach in this field is abstract argumentation frameworks developed by Dung. Assumption-based argumentation can be regarded as an instance of abstract argumentation with structured arguments. When facing a large scale of data, a challenge of reasoning in assumption-based argumentation is how to construct arguments and resolve attacks over a given claim with minimal cost of computation and acceptable accuracy at the same time. This thesis proposes and investigates approximate methods that randomly select and construct samples of frameworks based on graphical dispute derivations to solve this problem. The presented approach aims to improve reasoning performance and get an acceptable trade-off between computational time and accuracy. The evaluation shows that for reasoning in assumption-based argumentation, in general, the running time is reduced with the cost of slightly low accuracy by randomly sampling and constructing inference rules for potential arguments over a query.
Belief revision is the subarea of knowledge representation which studies the dynamics of epistemic states of an agent. In the classical AGM approach, contraction, as part of the belief revision, deals with the removal of beliefs in knowledge bases. This master's thesis presents the study and the implementation of concept contraction in the Description Logic EL. Concept contraction deals with the following situation. Given two concept C and D, assuming that C is subsumed by D, how can concept C be changed so that it is not subsumed by D anymore, but is as similar as possible to C? This approach of belief change is different from other related work because it deals with contraction in the level of concepts and not T-Boxes and A-Boxes in general. The main contribution of the thesis is the implementation of the concept contraction. The implementation provides insight into the complexity of contraction in EL, which is tractable since the main inference task in EL is also tractable. The implementation consists of the design of five algorithms that are necessary for concept contraction. The algorithms are described, illustrated with examples, and analyzed in terms of time complexity. Furthermore, we propose an new approach for a selection function, adapt for the concept contraction. The selection function uses metadata about the concepts in order to select the best from an input set. The metadata is modeled in a framework that we have designed, based on standard metadata frameworks. As an important part of the concept contraction, the selection function is responsible for selecting the best concepts that are as similar as possible to concept C. Lastly, we have successfully implemented the concept contraction in Python, and the results are promising.
Commonsense reasoning can be seen as a process of identifying dependencies amongst events and actions. Understanding the circumstances surrounding these events requires background knowledge with sufficient breadth to cover a wide variety of domains. In the recent decades, there has been a lot of work in extracting commonsense knowledge, a number of these projects provide their collected data as semantic networks such as ConceptNet and CausalNet. In this thesis, we attempt to undertake the Choice Of Plausible Alternatives (COPA) challenge, a problem set with 1000 questions written in multiple-choice format with a premise and two alternative choices for each question. Our approach differs from previous work by using shortest paths between concepts in a causal graph with the edge weight as causality metric. We use CausalNet as primary network and implement a few design choices to explore the strengths and drawbacks of this approach, and propose an extension using ConceptNet by leveraging its commonsense knowledge base.
Current political issues are often reflected in social media discussions, gathering politicians and voters on common platforms. As these can affect the public perception of politics, the inner dynamics and backgrounds of such debates are of great scientific interest. This thesis takes user generated messages from an up-to-date dataset of considerable relevance as Time Series, and applies a topic-based analysis of inspiration and agenda setting to it. The Institute for Web Science and Technologies of the University Koblenz-Landau has collected Twitter data generated beforehand by candidates of the European Parliament Election 2019. This work processes and analyzes the dataset for various properties, while focusing on the influence of politicians and media on online debates. An algorithm to cluster tweets into topical threads is introduced. Subsequently, Sequential Association Rules are mined, yielding wide array of potential influence relations between both actors and topics. The elaborated methodology can be configured with different parameters and is extensible in functionality and scope of application.
Our work finds the fine grained edits in context of neighbouring tokens in Wikipedia articles. We cluster those edits according to similar neighbouring context. We encode neighbouring context into vector space using word vectors. We evaluate clusters returned by our algorithm on extrinsic and intrinsic metric and compare it with previous work. We analyse the relation between extrinsic and intrinsic measurements of fine grained edit tokens.
Most social media platforms allow users to freely express their opinions, feelings, and beliefs. However, in recent years the growing propagation of hate speech, offensive language, racism and sexism on the social media outlets have drawn attention from individuals, companies, and researchers. Today, sexism both online and offline with different forms, including blatant, covert, and subtle lan- guage, is a common phenomenon in society. A notable amount of work has been done over identifying sexist content and computationally detecting sexism which exists online. Although previous efforts have mostly used peoples’ activities on social media platforms such as Twitter as a public and helpful source for collecting data, they neglect the fact that the method of gathering sexist tweets could be biased towards the initial search terms. Moreover, some forms of sexism could be missed since some tweets which contain offensive language could be misclassified as hate speech. Further, in existing hate speech corpora, sexist tweets mostly express hostile sexism, and to some degree, the other forms of sexism which also appear online was disregarded. Besides, the creation of labeled datasets with manual exertion, relying on users to report offensive comments with a tremendous effort by human annotators is not only a costly and time-consuming process, but it also raises the risk of involving discrimination under biased judgment.
This thesis generates a novel sexist and non-sexist dataset which is constructed via "UnSexistifyIt", an online web-based game that incentivizes the players to make minimal modifications to a sexist statement with the goal of turning it into a non-sexist statement and convincing other players that the modified statement is non-sexist. The game applies the methodology of "Game With A Purpose" to generate data as a side-effect of playing the game and also employs the gamification and crowdsourcing techniques to enhance non-game contexts. When voluntary participants play the game, they help to produce non-sexist statements which can reduce the cost of generating new corpus. This work explores how diverse individual beliefs concerning sexism are. Further, the result of this work highlights the impact of various linguistic features and content attributes regarding sexist language detection. Finally, this thesis could help to expand our understanding regarding the syntactic and semantic structure of sexist and non-sexist content and also provides insights to build a probabilistic classifier for single sentences into sexist or non-sexist classes and lastly find a potential ground truth for such a classifier.
Data visualization is an effective way to explore data. It helps people to get a valuable insight of the data by placing it in a visual context. However, choosing a good chart without prior knowledge in the area is not a trivial job. Users have to manually explore all possible visualizations and decide upon ones that reflect relevant and desired trend in the data, are insightful and easy to decode, have a clear focus and appealing appearance. To address these challenges we developed a Tool for Automatic Generation of Good viSualizations using Scoring (TAG²S²). The approach tackles the problem of identifying an appropriate metric for judging visualizations as good or bad. It consists of two modules: visualization detection: given a data-set it creates a list of combination of data attributes for scoring and visualization ranking: scores each chart and decides which ones are good or bad. For the later, an utility metric of ten criteria was developed and each visualization detected in the first module is evaluated on these criteria. Only those visualizations that received enough scores are then presented to the user. Additionally to these data parameters, the tool considers user perception regarding the choice of visual encoding when selecting a visualization. To evaluate the utility of the metric and the importance of each criteria, test cases were developed, executed and the results presented.
The content aggregator platform Reddit has established itself as one of the most popular websites in the world. However, scientific research on Reddit is hindered as Reddit allows (and even encourages) user anonymity, i.e., user profiles do not contain personal information such as the gender. Inferring the gender of users in large-scale could enable the analysis of gender-specific areas of interest, reactions to events, and behavioral patterns. In this direction, this thesis suggests a machine learning approach of estimating the gender of Reddit users. By exploiting specific conventions in parts of the website, we obtain a ground truth for more than 190 million comments of labeled users. This data is then used to train machine learning classifiers to use them to gain insights about the gender balance of particular subreddits and the platform in general. By comparing a variety of different approaches for classification algorithm, we find that character-level convolutional neural network achieves performance with an 82.3% F1 score on a task of predicting a gender of a user based on his/her comments. The score surpasses 85% mark for frequent users with more than 50 comments. Furthermore, we discover that female users are less active on Reddit platform, they write fewer comments and post in fewer subreddits on average, when compared to male users.
This Master Thesis is an exploratory research to determine whether it is feasible to construct a subjectivity lexicon using Wikipedia. The key hypothesis is that that all quotes in Wikipedia are subjective and all regular text are objective. The degree of subjectivity of a word, also known as ''Quote Score'' is determined based on the ratio of word frequency in quotations to its frequency outside quotations. The proportion of words in the English Wikipedia which are within quotations is found to be much smaller as compared to those which are not in quotes, resulting in a right-skewed distribution and low mean value of Quote Scores.
The methodology used to generate the subjectivity lexicon from text corpus in English Wikipedia is designed in such a way that it can be scaled and reused to produce similar subjectivity lexica of other languages. This is achieved by abstaining from domain and language-specific methods, apart from using only readily-available English dictionary packages to detect and exclude stopwords and non-English words in the Wikipedia text corpus.
The subjectivity lexicon generated from English Wikipedia is compared against other lexica; namely MPQA and SentiWordNet. It is found that words which are strongly subjective tend to have high Quote Scores in the subjectivity lexicon generated from English Wikipedia. There is a large observable difference between distribution of Quote Scores for words classified as strongly subjective versus distribution of Quote Scores for words classified as weakly subjective and objective. However, weakly subjective and objective words cannot be differentiated clearly based on Quote Score. In addition to that, a questionnaire is commissioned as an exploratory approach to investigate whether subjectivity lexicon generated from Wikipedia could be used to extend the coverage of words of existing lexica.
Ontologien sind wichtige Werkzeuge zur Wissensrepräsentation und elementare Bausteine des Semantic Web. Sie sind jedoch nicht statisch und können sich über die Zeit verändern. Die Gründe hierfür sind vielfältig: Konzepte innerhalb einer Ontologie können fehlerhaft modelliert worden sein, die von der Ontologie repräsentierte Domäne kann sich verändern oder eine Ontologie kann wiederverwendet werden und muss an den neuen Kontext angepasst oder mit bestehenden Ontologien verbunden werden. Die Schwierigkeit dieses Prozesses hat zur Entstehung des Forschungsfeldes der Ontology Change geführt. Das Entfernen von Wissen aus Ontologien ist ein wichtiger Aspekt dieses Änderungsprozesses, da selbst das Hinzufügen neuen Wissens zu einer Ontologie das Entfernen bestehenden Wissens notwendig machen kann, falls dieses mit den neuen Vorstellungen in Konflikt steht. Dieses Entfernen muss jedoch wohldurchdacht sein, da das Ändern bestehender Konzepte leicht zu viel Wissen aus der Ontologie entfernen oder die semantische Bedeutung der Konzepte auf eine potenziell unerwartete Weise verändern kann. In dieser Arbeit wird daher ein formaler Operator zum präzisen Entfernen von Wissen aus Konzepten vorgestellt. Dieser basiert auf der Beschreibungslogik EL und baut partiell auf den Postulaten für Belief Set und Belief Base Contraction sowie der Arbeit von Suchanek et al. auf. Hierfür wird zunächst ein Einstieg in das Thema Ontologien und die Ontologiesprache OWL 2 gegeben und das Problemfeld der Ontology Change wird erläutert. Es wird dann gezeigt, wie ein formaler Operator diesen Prozess unterstützen kann und weshalb die Beschreibungslogik EL einen guten Ausgangspunkt für die Entwicklung eines solchen Operators darstellt. Anschließend wird ein Einblick in das Feld der Beschreibungslogiken gegeben. Hierfür wird die Geschichte der Beschreibungslogik kurz umrissen, Anwendungsgebiete werden genannt und es werden Standardprobleme in dieser Logik erläutert. In diesem Zusammenhang wird die Beschreibungslogik EL formal eingeführt. In einem nächsten Schritt werden verwandte Arbeiten untersucht und es wird gezeigt, warum das Recovery- und Relevance-Postulat für das Entfernen von Wissen aus Konzepten nicht unmittelbar anwendbar ist. Die hier gewonnenen Erkenntnisse werden anschließend dazu genutzt, die Anforderungen an den Operator zu formalisieren. Diese basieren hauptsächlich auf den Postulaten für Belief Set und Belief Base Contraction. Zusätzlich werden weitere Eigenschaften formuliert welche den Verlust des Recovery- bzw. Relevance-Postulates ausgleichen sollen. In einem nächsten Schritt wird der Operator definiert und es wird gezeigt, dass diese Definition das präzise Entfernen von Wissen aus EL-Konzepten gestattet. Mittels formaler Beweise wird zudem gezeigt, dass diese Definition alle zuvor aufgestellten Anforderungen erfüllt. In einem weiteren Beispiel wird dargestellt, wie der Operator in Verbindung mit sogenannten Laconic Justifications verwendet werden kann, um einen menschlichen Ontology-Editor durch das automatisierte Entfernen von unerwünschten Konsequenzen aus der Ontologie zu unterstützen. Aufbauend auf Algorithmen, welche aus der formalen Definition des Operators abgeleitet wurden, wird ein Plugin zum Entfernen von Wissen aus Ontologien für den Ontology-Editor Protégé vorgestellt. Anschließend werden die bisherigen Erkenntnisse zusammengefasst und es wird ein Fazit gezogen. Die Arbeit schließt mit einem Ausblick über mögliche zukünftige Forschung.
We examine the systematic underrecognition of female scientists (Matilda effect) by exploring the citation network of papers published in the American Physical Society (APS) journals. Our analysis shows that articles written by men (first author, last author and dominant gender of authors) receive more citations than similar articles written by women (first author, last author and dominant gender of authors) after controlling for the journal of publication, year of publication and content of the publication. Statistical significance of the overlap between the lists of references was considered as the measure of similarity between articles in our analysis. In addition, we found that men are less likely to cite articles written by women and women are less likely to cite articles written by men. This pattern leads to receiving more citations by articles written by men than similar articles written by women because the majority of authors who published in APS journals are male (85%). We also observed Matilda effect reduces when articles are published in journals with the highest impact factors. In other words, people’s evaluation of articles published in these journals is not affected by the gender of authors significantly. Finally, we suggested a method that can be applied by editors in academic journals to reduce the evaluation bias to some extent. Editors can identify missing citations using our proposed method to complete bibliographies. This policy can reduce the evaluation bias because we observed papers written by female scholars (first author, last author, the dominant gender of authors) miss more citations than articles written by male scholars (first author, last author, the dominant gender of authors).
Knowledge-based authentication methods are vulnerable to Shoulder surfing phenomenon.
The widespread usage of these methods and not addressing the limitations it has could result in the user’s information to be compromised. User authentication method ought to be effortless to use and efficient, nevertheless secure.
The problem that we face concerning the security of PIN (Personal Identification Number) or password entry is shoulder surfing, in which a direct or indirect malicious observer could identify the user sensitive information. To tackle this issue we present TouchGaze which combines gaze signals and touch capabilities, as an input method for entering user’s credentials. Gaze signals will be primarily used to enhance targeting and touch for selecting. In this work, we have designed three different PIN entry method which they all have similar interfaces. For the evaluation, these methods were compared based on efficiency, accuracy, and usability. The results uncovered that despite the fact that gaze-based methods require extra time for the user to get familiar with yet it is considered more secure. In regards to efficiency, it has the similar error margin to the traditional PIN entry methods.
Topic Models sind ein beliebtes Werkzeug um Themen in großen Textkorpora zu identifizieren. Diese Textkorpora enthalten oft versteckte Meta-Gruppen. Das Größenverhältnis zwischen diesen Gruppen variiert meist stark. Die Präsenz dieser Gruppen wird in der Praxis oft ignoriert. Diese Masterarbeit erforscht daher, ob diese Gruppen Einfluss auf ein Topic Model haben.
Um den Einfluss zu testen, wird LDA auf Samples mit unterschiedlichen Gruppengrößen trainiert. Die Samples werden von Textkorpora mit großen Gruppenunterschieden (d.h. Sprachunterschieden) und kleinen Gruppenunterschieden (d.h. Unterschiede in der politische Orientierung) generiert. Die Leistungsfähigkeit von LDA wird per "Perplexity" evaluiert.
Der Einfluss von Gruppen auf die generelle Leistungsfähigkeit von Topic Models hängt von verschiedenen Faktoren der Gruppen ab, z.B. der Vorhersagbarkeit der Sprache generell. Die Leistungsfähigkeit der Topic Models für die einzelnen Gruppen wird von der Variation der relativen Gruppengrößen beeinflusst. Allerdings ist der Effekt für alle Datensätze verschieden.
LDA kann die Gruppen intern unterscheiden, wenn die Unterschiede der Gruppen groß genug sind (z.B. Sprachunterschiede). Der Anteil der Topics, die explizit für eine Gruppe gelernt werden, ist jedoch unterproportional zu dem Anteil der Gruppe im Trainingskorpus. Dieser Effekt verstärkt sich für kleinere Minderheiten.
The purpose of this thesis is to explore the sentiment distributions of Wikipedia concepts.
We analyse the sentiment of the entire English Wikipedia corpus, which includes 5,669,867 articles and 1,906,375 talks, by using a lexicon-based method with four different lexicons.
Also, we explore the sentiment distributions from a time perspective using the sentiment scores obtained from our selected corpus. The results obtained have been compared not only between articles and talks but also among four lexicons: OL, MPQA, LIWC, and ANEW.
Our findings show that among the four lexicons, MPQA has the highest sensitivity and ANEW has the lowest sensitivity to emotional expressions. Wikipedia articles show more sentiments than talks according to OL, MPQA, and LIWC, whereas Wikipedia talks show more sentiments than articles according to ANEW. Besides, the sentiment has a trend regarding time series, and each lexicon has its own bias regarding text describing different things.
Moreover, our research provides three interactive widgets for visualising sentiment distributions for Wikipedia concepts regarding the time and geolocation attributes of concepts.
The output of eye tracking Web usability studies can be visualized to the analysts as screenshots of the Web pages with their gaze data. However, the screenshot visualizations are found to be corrupted whenever there are recorded fixations on fixed Web page elements on different scroll positions. The gaze data are not gathered on their fixated fixed elements; rather they are scattered on their recorded scroll positions. This problem has raised our attention to find an approach to link gaze data to their intended fixed elements and gather them in one position on the screenshot. The approach builds upon the concept of creating the screenshot during the recording session, where images of the viewport are captured on visited scroll positions and lastly stitched into one Web page screenshot. Additionally, the fixed elements in the Web page are identified and linked to their fixations. For the evaluation, we compared the interpretation of our enhanced screenshot against the video visualization, which overcomes the problem. The results revealed that both visualizations equally deliver accurate interpretations. However, interpreting the visualizations of eye tracking Web usability studies using the enhanced screenshots outperforms the video visualizations in terms of speed and it requires less temporal demands from the interpreters.
“Did I say something wrong?” A word-level analysis of Wikipedia articles for deletion discussions
(2016)
Diese Arbeit beschäftigt sich damit, linguistische Erkenntnisse auf Wortebene über schriftlichen Diskussionen zu gewinnen. Die Unterscheidung zwischen Botschaften, welche sich förderlich auf Diskussionen auswirken und jene, welche diese unterbrechen, spielte dabei eine besondere Rolle. Hierbei lag ein Schwerpunkt darauf, zu ermitteln, ob Ich- und Du-Botschaften charakteristisch für die beiden Kommunikationsarten sind. Diese Botschaften sind über Jahre hinweg zu Empfehlungen für erfolgreiche Kommunikation avanciert. Ihre zugeschriebene Wirkung wurde zwar mehrfach bestätigt, jedoch geschah dies stets in kleineren Studien. Deshalb wurde in dieser Arbeit mithilfe der Löschdiskussionen der englischen Wikipedia und der Liste gesperrter Nutzer eine vollautomatische Erstellung eines annotierten Datensatzes entwickelt. Dabei wurden Diskussionsbotschaften entweder als förderlich oder schädlich für einen konstruktiven Diskussionsverlauf markiert. Dieser Datensatz wurde anschließend im Rahmen einer binären Klassifikation verwendet, um charakteristische Worte für die beiden Kommunikationsarten zu bestimmen. Es wurde zudem untersucht, ob anhand von Synsemantika (auch bekannt als Funktionswörter) wie Pronomen oder Konjunktionen eine Entscheidung über die Kommunikationsart einer Botschaft getroffen werden kann. Du-Botschaften wurden, übereinstimmend mit ihrer zugeschriebenen negativen Auswirkung auf Kommunikation, als schädlich in den durchgeführten Untersuchungen identifiziert. Entgegen der zugeschriebenen positiven Auswirkung von Ich-Botschaften, wurde bei diesen ebenfalls eine schädlich Wirkung festgestellt. Eine klare Aussage über die Relevanz von Synsemantika konnte anhand der Ergebnisse nicht getroffen werden. Weitere charakteristische Worte konnten nicht festgestellt werden. Die Ergebnisse deuten darauf hin, dass ein anderes Modell textliche Diskussionen potentiell besser abbilden könnte.
Politische und gesellschaftliche Polarisierung ist ein interessantes Phänomen, über dessen Auswirkungen viele unterschiedliche, zum Teil auch gegensätzliche, Theorien existieren.
Polarisierung wird in der Literatur mit unterschiedlichen Methoden gemessen. Die vorliegende Arbeit gibt einen Überblick über existierende Polarisierungsmaße und es werden zwei neuartige Maße aus dem Gebiet der spektralen Graphentheorie vorgestellt. Anschließend werden die bekannten und die neu entwickelten Maße auf den LiquidFeedback-Datensatz der Piratenpartei Deutschland angewandt. Als Ergebnis lässt sich festhalten, dass die Maße teilweise zu unterschiedlichen Ergebnisse kommen. Dies liegt darin begründet, dass nicht alle Maße das Gleiche messen. Um zu verstehen was die einzelnen Maße aussagen, werden wesentliche Eigenschaften von Polarisierungsmaßen herausgearbeitet und es wird für jedes Maß dargelegt, welche Eigenschaften es erfüllt. Die angesprochenen Polarisierungsmaße beziehen sich auf die Entwicklung der Polarisierung zwischen Usern des LiquidFeedback-Systems. Bei der Betrachtung von einzelnen Personen und Abstimmungen fiel unter anderem auf, dass polarisierende Personen mehr Macht durch
Delegationen besitzen als die restlichen Personen und dass polarisierte Vorschläge circa doppelt so häufig umgesetzt werden.