English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

English. This paper describes my attempt to build an artificial player for a very popular language game, called “The Guillotine”, within the Evalita Challenge (Basile et al., 2018). I have built this artificial player to investigate how far we can go by using resources available on the web and a simple matching algorithm. The resources used are Morph-it (Zanchetta and Baroni, 2005) and other online resources. The resolution algorithm is based on two steps: in the first step, it interrogates the knowledge base Morph-it with the five data clues, download the results and perform various intersection operations between the five data sets; in the second step, it refines the results through the other sources such as the Italian proverbs database and the IMDb. My artificial player identified the solution among the first 100 solutions proposed in 25% of cases. This is still far from systems like OTTHO (Semeraro et al., 2012) that obtained the solution in 68% of the cases. However, their result was obtained larger resources and not only with a simple web analysis. Italiano. Il contributo descrive il tentativo di costruire un giocatore artificiale per un gioco linguistico molto popolare, chiamato ”La Ghigliottina”, nell’ambito dell’Evalita Challenge (Basile et al., 2018). Ho costruito questo giocatore artificiale per verificare il limite raggiungibile utilizzando unicamente le risorse disponibili sul web e un semplice algoritmo di matching. Le risorse utilizzate sono Morph-it (Zanchetta and Baroni, 2005) e altre risorse online. L’algoritmo di risoluzione si basa su due fasi: nella prima fase, interroga la base di conoscenza Morph-it con i cinque indizi, scarica i risultati ed esegue varie operazioni di intersezione tra i cinque set di dati; nella seconda fase, affina i risultati attraverso altre fonti come il database dei proverbi italiani e l’IMDb. Il mio giocatore artificiale ha identificato la soluzione tra le prime 100 soluzioni proposte nel 25% dei casi. Il risultato ottenuto è ancora lontano da sistemi come OTTHO (Semeraro et al., 2012) che ha ottenuto la soluzione nel 68% dei casi. Tuttavia, il loro risultato è stato ottenuto con risorse più ampie e non solo con una semplice analisi web. 1 System description I have used Morph-it (Zanchetta and Baroni, 2005) as a basis for knowledge, instead of building one, as it is free, easy to interrogate and above all suitable for our purpose. Furthermore, it should not be underestimated that building a knowledge base involves an enormous amount of work in terms of time. After querying the knowledge base with the five data clues, the results are downloaded and various intersection operations are performed between the five data sets. This procedure allows us to find all possible solutions and is the basis for choosing the solution. Then to find the final solution is verified the existence of proverbs, Aphorisms, movies or books (etc.) that contain both the clue and the possible solution. 2 The memory of the system: Morph-it! Morph-it! is a free morphological resource for the Italian language, a lexicon of inflected forms with their lemma and morphological features. It was

Computer challenges guillotine: how an artificial player can solve a complex language TV game with web data analysis