Intelligent agents have the potential to understand personality traits of human beings because of their every day interaction with us. The assessment of our psychological traits is a useful tool when we require them to simulate empathy. Since the creation of social media platforms, numerous studies dealt with measuring personality traits by gathering users’ information from their social media profiles. Real world applications showed how natural language processing combined with supervised machine learning algorithms are effective in this field. These applications have some limitations such as focusing on English text only and not considering polysemy in text. In this paper, we propose a multilingual model that handles polysemy by analyzing sentences as a semantic ensemble of interconnected words. The proposed approach processes Facebook posts from the myPersonality dataset and it turns them into a high-dimensional array of features, which are then exploited by a deep neural network architecture based on transformer to perform regression. We prove the effectiveness of our work by comparing the mean squared error of our model with existing baselines and the Kullback–Leibler divergence between the relative data distributions. We obtained state-of-the-art results in personality traits estimation from social media posts for all five personality traits.

In social media, users are spreading misinformation easily and without fact checking. In principle, they do not have a malicious intent, but their sharing leads to a socially dangerous diffusion mechanism. The motivations behind this behavior have been linked to a wide variety of social and personal outcomes, but these users are not easily identified. The existing solutions show how the analysis of linguistic signals in social media posts combined with the exploration of network topologies are effective in this field. These applications have some limitations such as focusing solely on the fake news shared and not understanding the typology of the user spreading them. In this paper, we propose a computational approach to extract features from the social media posts of these users to recognize who is a fake news spreader for a given topic. Thanks to the CoAID dataset, we start the analysis with 300 K users engaged on an online micro-blogging platform; then, we enriched the dataset by extending it to a collection of more than 1 M share actions and their associated posts on the platform. The proposed approach processes a batch of Twitter posts authored by users of the CoAID dataset and turns them into a high-dimensional matrix of features, which are then exploited by a deep neural network architecture based on transformers to perform user classification. We prove the effectiveness of our work by comparing the precision, recall, and $f_1$ score of our model with different configurations and with a baseline classifier. We obtained an $f_1$ score of 0.8076, obtaining an improvement from the state-of-the-art by 4%.

Several techniques exist for test automation in the mobile domain, from script-based techniques to automated test generation based on GUI models. Most techniques, however, fall short in being adopted extensively by practitioners because of the very high cost for the definition (and maintainence) of test cases. Random testing tools decouple the test generation from the specific app (or version) to test, however they create unreproducible tests often with a low bug-finding ability. The objective of this work is to build a testing framework for Android apps, capable of allowing the user to write very simple test scripts without knowing details about the application implementation details and appearance. The main requirement for those tests is to generate adaptive tests that can be executed on a great number of apps without manual re-adaptations. We developed a testing framework that: (1) models the application by means of state graphs; (2) by using machine learning algorithms, classifies the type of application under test and each activity (i.e., screen) of the application; (3) allows to create general test cases, specific to application and activity categories, that are then automatically adapted to the AUT without effort from the tester. We created two separated classifiers (for application and activity type) that outperform available approaches in literature. On top of them, the developed testing framework was able to perfectly adapt high-level test cases to 28 of 32 tested applications. Such high-level tests allow a reduction of LOCs of the test suite of around 90%. Our findings prove that a machine learning-based approach can be derived for the creation of high-level, adaptive test cases for different typologies of Android applications and activities. Our prototype framework is modular in nature, and allows expansions by means of addition of new commands to be executed on the classified apps and activities.

Micro-influencers have triggered the interest of commercial brands, public administrations, and other stakeholders because of their demonstrated capability of sensitizing people within their close reach. However, due to their lower visibility in social media platforms, they are challenging to be identified. This work proposes an approach to automatically detect micro-influencers and to highlight their personality traits and community values by computationally analyzing their writings. We introduce two learning methods to retrieve Five Factor Model and Basic Human Values scores. These scores are then used as feature vectors of a Support Vector Machines classifier. We define a set of rules to create a micro-influencer gold standard dataset of more than two million tweets and we compare our approach with three baseline classifiers. The experimental results favor recall meaning that the approach is inclusive in the identification.

This article reports on the XML retrieval system x² that has been developed at the University of Munich over the last 5 years. In a typical session with x², the user first browses a structural summary of the XML database in order to select interesting elements and keywords occurring in documents. Using this intermediate result, queries combining structure and textual references are composed semiautomatically. After query evaluation, the full set of answers is presented in a visual and structured way. x² largely exploits the structure found in documents, queries and answers to enable new interactive visualization and exploration techniques that support mixed IR and database-oriented querying, thus bridging the gap between these three views on the data to be retrieved. Another salient characteristic of x² that distinguishes it from other visual query systems for XML is that it supports various degrees of detailedness in the presentation of answers, as well as techniques for dynamically reordering, grouping and ranking retrieved elements once the complete answer set has been computed.