Workshop on Games and Natural Language Processing
Stephanie M. Lukin (Editor)
A weak point of rule-based sentiment analysis systems is that the underlying sentiment lexicons are often not adapted to the domain of the text we want to analyze. We created a game-specific sentiment lexicon for video game Skyrim based on the E-ANEW word list and a dataset of Skyrim’s in-game documents. We calculated sentiment ratings for NPC dialogue using both our lexicon and E-ANEW and compared the resulting sentiment ratings to those of human raters. Both lexicons perform comparably well on our evaluation dialogues, but the game-specific extension performs slightly better on the dominance dimension for dialogue segments and the arousal dimension for full dialogues. To our knowledge, this is the first time that a sentiment analysis lexicon has been adapted to the video game domain.
The ESP Game (also known as the Google Image Labeler) demonstrated how the crowd could perform a task that is straightforward for humans but challenging for computers – providing labels for images. The game facilitated the task of basic image labeling; however, the labels generated were non-specific and limited the ability to distinguish similar images from one another, limiting its ability in search tasks, annotating images for the visually impaired, and training computer vision machine algorithms. In this paper, we describe ClueMeIn, an entertaining web-based game with a purpose that generates more detailed image labels than the ESP Game. We conduct experiments to generate specific image labels, show how the results can lead to improvements in the accuracy of image searches over image labels generated by the ESP Game when using the same public dataset.
Errors commonly exist in machine-generated documents and publication materials; however, some correction algorithms do not perform well for complex errors and it is costly to employ humans to do the task. To solve the problem, a prototype computer game called Cipher was developed that encourages people to identify errors in text. Gamification is achieved by introducing the idea of steganography as the entertaining game element. People play the game for entertainment while they make valuable annotations to locate text errors. The prototype was tested by 35 players in a evaluation experiment, creating 4,764 annotations. After filtering the data, the system detected manually introduced text errors and also genuine errors in the texts that were not noticed when they were introduced into the game.
GWAP design might have a tremendous effect on its popularity of course but also on the quality of the data collected. In this paper, a comparison is undertaken between two GWAPs for building term association lists, namely JeuxDeMots and Quicky Goose. After comparing both game designs, the Cohen kappa of associative lists in various configurations is computed in order to assess likeness and differences of the data they provide.
In this paper, we describe a Telegram bot, Mago della Ghigliottina (Ghigliottina Wizard), able to solve La Ghigliottina game (The Guillotine), the final game of the Italian TV quiz show L’Eredità. Our system relies on linguistic resources and artificial intelligence and achieves better results than human players (and competitors of L’Eredità too). In addition to solving a game, Mago della Ghigliottina can also generate new game instances and challenge the users to match the solution.
Gamification has been applied to many linguistic annotation tasks, as an alternative to crowdsourcing platforms to collect annotated data in an inexpensive way. However, we think that still much has to be explored. Games with a Purpose (GWAPs) tend to lack important elements that we commonly see in commercial games, such as 2D and 3D worlds or a story. Making GWAPs more similar to full-fledged video games in order to involve users more easily and increase dissemination is a demanding yet interesting ground to explore. In this paper we present a 3D role-playing game for abusive language annotation that is currently under development.
In this paper we present a new method for collecting naturally generated dialogue data for a low resourced language, (specifically here—Uyghur). We plan to build a games with a purpose (GWAPs) to encourage native speakers to actively contribute dialogue data to our research project. Since we aim to characterize the response space of queries in Uyghur, we design various scenarios for conversations that yield to questions being posed and responded to. We will implement the GWAP with the RPG Maker MV Game Engine, and will integrate the chatroom system in the game with the Dialogue Experimental Toolkit (DiET). DiET will help us improve the data collection process, and most importantly, make us have some control over the interactions among the participants.
In this paper, we present the ongoing development of CALLIG – a web system that uses improvisation games in Computer Assisted Language Learning (CALL). Improvisation games are structured activities with built-in constraints where improvisers are asked to generate a lot of different ideas and weave a diverse range of elements into a sensible narrative spontaneously. This paper discusses how computer-based language games can be created combining improvisation elements and language technology. In contrast with traditional language exercises, improvisational language games are open and unpredictable. CALLIG encourages spontaneity and witty language use. It also provides opportunities for collecting useful data for many NLP applications.
Although the roguelike video game genre has a large community of fans (both players and developers) and the graphic aspect of these games is usually given little relevance (ASCII-based graphics are not rare even today), their accessibility for blind players and other visually-impaired users remains a pending issue. In this document, we describe an initiative for the development of roguelikes adapted to visually-impaired players by using Natural Language Processing techniques, together with the first completed games resulting from it. These games were developed as Bachelor’s and Master’s theses. Our approach consists in integrating a multilingual module that, apart from the classic ASCII-based graphical interface, automatically generates text descriptions of what is happening within the game. The visually-impaired user can then read such descriptions by means of a screen reader. In these projects we seek expressivity and variety in the descriptions, so we can offer the users a fun roguelike experience that does not sacrifice any of the key characteristics that define the genre. Moreover, we intend to make these projects easy to extend to other languages, thus avoiding costly and complex solutions. KEYWORDS: Natural Language Generation, roguelikes, visually-impaired users
Increasing efforts are put into gamification of experimentation software in psychology and educational applications and the development of serious games. Computer-based experiments with game-like features have been developed previously for research on cognitive skills, cognitive processing speed, working memory, attention, learning, problem solving, group behavior and other phenomena. It has been argued that computer game experiments are superior to traditional computerized experimentation methods in laboratory tasks in that they represent holistic, meaningful, and natural human activity. We present a novel experimental framework for forced choice categorization tasks or speech perception studies in the form of a computer game, based on the Unity Engine – the Gamified Discrimination Experiments engine (GDX). The setting is that of a first person shooter game with the narrative background of an alien invasion on earth. We demonstrate the utility of our game as a research tool with an application focusing on attention to fine phonetic detail in natural speech perception. The game-based framework is additionally compared against a traditional experimental setup in an auditory discrimination task. Applications of this novel game-based framework are multifarious within studies on all aspects of spoken language perception.
As the uses of Games-With-A-Purpose (GWAPs) broadens, the systems that incorporate its usages have expanded in complexity. The types of annotations required within the NLP paradigm set such an example, where tasks can involve varying complexity of annotations. Assigning more complex tasks to more skilled players through a progression mechanism can achieve higher accuracy in the collected data while acting as a motivating factor that rewards the more skilled players. In this paper, we present the progression technique implemented in Wormingo , an NLP GWAP that currently includes two layers of task complexity. For the experiment, we have implemented four different progression scenarios on 192 players and compared the accuracy and engagement achieved with each scenario.
Automatic Annotation of Werewolf Game Corpus with Players Revealing Oneselves as Seer/Medium and Divination/Medium Results
Youchao Lin | Miho Kasamatsu | Tengyang Chen | Takuya Fujita | Huanjin Deng | Takehito Utsuro
While playing the communication game “Are You a Werewolf”, a player always guesses other players’ roles through discussions, based on his own role and other players’ crucial utterances. The underlying goal of this paper is to construct an agent that can analyze the participating players’ utterances and play the werewolf game as if it is a human. For a step of this underlying goal, this paper studies how to accumulate werewolf game log data annotated with identification of players revealing oneselves as seer/medium, the acts of the divination and the medium and declaring the results of the divination and the medium. In this paper, we divide the whole task into four sub tasks and apply CNN/SVM classifiers to each sub task and evaluate their performance.