 Hello everyone, my name is Dylan Baumgartner, and I am a bachelor's student at the University of Zurich in Switzerland and I will be presenting expert or research relating to my bachelor thesis work where we are trying to study the extent to which the discourse quality index can be applicable in Wikipedia Deliberation and at the same time we are trying to better understand the current wiki deliberative processes and Factors that are related to the quality the deliberation While in our extended abstract we presented the mythology of our work. We already have some early results that we would like to share Our motivation for this work lies within how deliberation is an essential element of the peer production process in wiki media projects furthermore recent Literature has looked into the phenomenon of how wikipedia rcs becomes stale and also how group diversity can influence the wiki data knowledge graph Social psychology has looked into how diverse groups can be effective. However Must be managed effectively in order to produce Good outcomes While political science research focused mainly on the relation between gender diversity and ethnicity diversity and Deliberation our work studies deliver quality based on a political science standard the discourse quality index Our goal is twofold firstly to compare the deliberative quality of discussions taking place in rcs of the English wikipedia wiki data and meta and Secondly to study the relationship between group diversity and deliberative quality Following our motivation. We stated the following research questions How does the deliberative quality in wikipedia wiki data and metas rcs discussions compare? How does the lifespan of the rcs discussions compare across the three projects? Relating to the participants. How does the diversity compare across all projects? And then finally, what is to what extent is the participants diversity related to the deliberative quality of rcs? in order to Answer our research questions. We collected data Using py wiki bots more specifically we collected pages referencing specific templates so in wikipedia the Request for comments are embedded in talk pages and we then looked for the talk pages referencing a Specific template called closed rc top once we had that page. We then parsed it the wiki text until closed rc bottom for example and in wiki data and met to rcs the RFCs are an individual pages or is it easier and we just had a look for a specific template We've been following parsed the discussions in the talk pages and then converted the templates in wiki text to human readable text We collected available user information the wiki age at account number of wiki projects with more than 50 edits and The roles via the wiki media API We focused on closed rcs as these relate closely to real parliamentary debates where there's any end outcome and Not becoming stale Relating to the content of the RSC so in wikipedia the RSCs are mostly concerned with articles sources in wiki data about the knowledge graph and in wiki media the community users and blocking So we use the discourse quality index It's defined in the political science domain and it spans over the following dimensions however, after labeling 200 comments we found only the metrics colored to be applicable for our data set and The metrics in red were trained with a random forest classification a supervised learning method and Explanation didn't work well Using the random forest classification and we used a legit logistic regression method based on text length and And about a 75 precision rate that way disrespect was not Available in our data sets wasn't that there wasn't that many comments and We measured it as toxicity via a Burt's based model and our scale is then from one to five We tested for normality using the Kolmogorov-Semyonov test and found our The deliberate quality to not be normally distributed We analyzed the differences between diversity means using the man-with-me-you test and Found a significant difference between all pairs of projects We found meta and wiki data RSEs to be open for a longer time We measured the participants diversity or experience based on the following four metrics world diversity age diversity edit diversity and the amount of projects with over 50 edits We calculated the diversity index using the Shannon entropy and we also found the distribution to follow a non-normal distribution and Also found a significant difference across all projects upon investigating if there's a correlation between the participants diversity and Deliberative quality we found a slight correlation within wikipedia RSEs and a bit less but also in meta so this was a first step towards observing understanding and Possibly improving deliberative quality in terms of the DQI in wiki media projects We found that deliberative quality was around the same values in all projects, but with slightly different distributions concluding we found meta and wiki data RSEs to be open for a longer time we found a small positive correlation between some group diversities and deliberation quality within wikipedia in meta RSEs and We also found that deliberation is often within voting sections and We would propose To have a separate voting section maybe with a graphical user interface where a user can Maybe click on a support button instead of loading the Discussion sections with support oppose comment, etc Thank you for attention and if you have any questions now will be the time to do so. Thank you