English. Code-mixing is the alternation between two or more languages in the same text. This phenomenon is very relevant in the travel domain, since it can provide new insight in the way foreign cultures are perceived and described to the readers. In this paper, we analyse EnglishItalian code-mixing in historical English travel writings about Italy. We retrain and compare two existing systems for the automatic detection of code-mixing, and analyse the semantic categories mostly connected to Italian. Besides, we release the domain corpus used in our experiments and the output of the extraction.
Italiano. Il code-mixing e` l’alternanza di lingue diverse nello stesso testo. Questo fenomeno e` particolarmente importante nel dominio dei viaggi, poiche´ aiuta a comprendere meglio il modo in cui vengono percepite e descritte culture diverse da quella dell’autore. In questo lavoro, analizziamo il code-mixing tra inglese ed italiano nei testi di viaggio scritti in inglese e aventi come soggetto l’Italia. A questo scopo confrontiamo due sistemi esistenti per il riconoscimento automatico del code-mixing dopo averli ri-addestrati e analizziamo le categorie semantiche connesse alle parole/espressioni italiane. Inoltre, rilasciamo il corpus e il risultato dell’estrazione.
Code-mixing is the alternation between two or
more languages that can occur between sentences
(inter-sentential), within the same utterance
(intrasentential), or even inside a single token
(mixing of morphemes). This phenomenon has been
widely studied from the linguistic,
psycholinguistic, and sociolinguistic point of view
Code-mixing characterizes communication of
post-colonial, migrant and multilingual
communities
In this work, we investigate the phenomenon of code-mixing in travel writings, but differently from previous works we shift the focus of analysis from contemporary to historical data and from manual to automatic information extraction. As for the first point, we present a corpus of more than 3.5 millions words of English travel writings published between the end of the XIX Century and the beginning of the XX Century, which we have retrieved from freely available sources and we release in a cleaned format. As for automatic information extraction, we retrain two state-of-the-art tools to identify English-Italian code-mixing and evaluate them on a sample of our dataset. We further launch the best system on the whole dataset and then we perform a semi-automatic refinement of the automatic annotation. The corpus, the training and test data and the outcome of the extraction are available online1. 2
Automatic language identification of monolingual
documents has a long tradition in Natural
Language Processing
To the best our knowledge, there is no
previous work on the automatic identification of
codemixing in travel writing. Cappelli (2013) and
Gandin (2014) have studied the phenomenon, but
they have mainly used standard corpus
linguistics tools, i.e. WordSmith
Differently from the works cited in the
previous Section, we focus on historical texts. To
this end, we collect from Project Gutenberg2 a
corpus of travel writings about Italy written by
English native authors and published between
the country unification and the beginning of the
30’s. We choose this period because in the
second half of the XIX Century the tradition of the
Grand Tour declined and leisure-oriented travels
emerged. This radical transformation was
enabled by technological, economic and
sociological, factors, such as the development of
steampowered ships and of the railway network, the
1https://dh.fbk.eu/technologies/
code-mixing
2https://www.gutenberg.org/
growth of Anglo-American economy and a greater
emancipation of women with more female
travelers
The corpus is made by 57 texts3, divided into
travel narratives (reports, diaries, collections of
letters) and guidebooks, for a total of 3,630,781
tokens. We distinguish between these two types
of text, following a standard classification of
documents in the travel domain. However, the
distinction was not so clear-cut in the period we take
into account as it is now, since reports on
personal travel experiences were often mixed with
practical recommendations and long disquisitions
on art and history. Therefore, we adopt as a rule
of thumb the distinction suggested in
The authors of the selected texts belong to different nationalities (UK, US, Ireland, Australia) and are both male and female. Some books dwell on specific cities or regions, others cover different parts of Italy or even several countries: in the latter case we extracted only the chapters related to Italy. Although we made an effort to have a diverse, well-balanced corpus in terms of content, author’s gender and nationality, this was only partially possible because of the limited availability of online travel books whose text is freely available and cleaned from OCR errors. The distribution of tokens according to the year of publication and type of text is shown in Fig. 1. Details about authors are given in a spreadsheet provided together with the corpus. 4
In this Section we describe the experiments on code-mixing, comparing the performance of two available systems in different configurations. We also detail the post-processing step introduced to refine the output of the best performing system.
3Thirty of these texts are also available in TEI-XML
format on the website https://sites.google.com/
view/travelwritingsonitaly.
In order to automatically extract Italian words,
expressions and sentences from the corpus
described in Section 3, we train and test two systems
whose source code is available on the web. The
first one (henceforth, langid) is based on
character n-grams (n = 1 to 5) and adopts a weakly
supervised approach, i.e. training data are
monolingual texts of few thousand tokens
For our experiments, we retrain langid using a collection of about 300,000 tokens taken from monolingual Italian and English books, of different genres, published in the same period of our corpus5.
The second system (henceforth, CodeSwitching), has been developed to detect languages in texts mixing Latin and Middle English (Schulz
˜be5FnokrinItgal/iarne: s“oLuercAevsve/nltuarnegdiidP_irneoclcehaios”eb.ytaCr..Cgozllodi, “Una donna” by S. Aleramo, “Il Valdarno da Firenze al mare” by G. Carocci, “La vita operosa” by M. Bontempelli, “Dopo il divorzio” by G. Deledda, “Novelle umoristiche” by A. Albertazzi, “Lezioni e Racconti per i bambini” by I. Baccini. For English: “The Adventures of Tom Sawyer” by M. Twain, “Pioneers of the Old Southwest” by C. L. Skinner, “The Happy Prince, and Other Tales” by O. Wilde, “Vanished Arizona” by M. Summerhayes, “The Tale of Peter Rabbit” by B. Potter, “The Strange Case of Dr. Jekyll and Mr. Hyde” by R. L. Stevenson. and Keller, 2016). It implements a CRF classifier with features generated from TreeTagger models and word lists of both languages6. Differently from langid that classifies words as belonging to one language rather than the other, this latter system performs a fine-grained annotation by distinguishing five classes (see below). Since this system is fully supervised, we create a training set by manually annotating 3,900 tokens from 4 samples extracted from our corpus, a size in line with the training data used in the original paper. The training data were annotated with 5 different classes: Italian tokens (i), English tokens (e), punctuation (p), named entities (NEs) (n), and ambiguous tokens that belong to the dictionary of both languages (a).
Both langid and CodeSwitching were evaluated
on the same test set, i.e. two samples of texts
(one from a travel narrative and one from a
guidebook) of 1,623 tokens. The test set was
annotated by assigning to each token a label for English
or Italian, as required by langid, and also
marking punctuation, NEs and ambiguous tokens,
following CodeSwitching scheme. Since the
performance of CodeSwitching is sensitive to the length
of the input file, we split the test set in batches of
40 sentences, replicating the experimental setting
presented in
Table 1 presents the performances of langid on the test set: contrary to the results achieved by King and Abney (2013), HMM – not CRF – proved to be the best approach. This is likely due to the greater sparseness of the code-mixing phenomenon in our dataset with respect to what was registered in the original corpus, where languages different from English cover the 56% of the overall number of tokens.
Table 2 reports Precision, Recall and F-measure of the retrained CodeSwitching system. Even if the overall performance is slightly better than the one obtained with HMM in langid, the scores for the detection of Italian tokens (i) are lower (0.82 versus 0.90 in terms of F-measure). Punctuation (i) and ambiguous tokens (a) are generally detected with a good performance, while NEs (e) represent the most challenging class. Given that we are mainly interested in recognising English and Ital
CodeSwitching ian terms, and that on this task langid performs better, we run this tool on the whole corpus. 4.3
In order to refine the output of langid (see Figure
2), we perform three post-processing steps. First
of all, we check whether tokens tagged as Italian
are included in Morph-it, an Italian lexicon of
inflected forms
The classification performed with the USAS tagger shows that Italian is adopted to express con
latest/Installation.html
cepts covered by 20 semantic classes, both in
guidebooks and in travel narratives. Only one
USAS class, the one related to “Science and
technology”, is not found in the corpus. Table 5 shows
frequency and examples for each detected class.
As in contemporary travel writings
In many cases, the use of Italian is not limited to single words or multi-token expressions (e.g. appartamento signorile) but longer utterances are reported. Texts of both genres contain proverbs (e.g. chi tardi arriva mal alloggia) and citations, not only from the canon of Italian literature, such as Leopardi’s poems, but also from the popular tradition, such as Tuscan songs (O rosa O rosa O rosa gentillina). The main difference between travel narratives and guidebooks is the greater presence in the former of dialogues or expressions heard by the author during his/her stay in Italy (voi siete un
SEMANTIC CLASS # names & grammar 29,927 architecture 3,070 movement 2,294 social elements 1,590 materials & objects 717 environment 713 general/abstract terms 580 measurement 340 arts & crafts 231 time 225 life 222 body 211 public domain 205 psyche 198 food & farming 162 entertainment 141 emotion 137 communication 131 money & commerce 127 education 22
SEMANTIC CLASS # EXAMPLES names & grammar 28,694 Donatello social elements 3,134 popolo architecture 3,065 palazzo environment 1,311 lago movement 1,207 vetturino materials & objects 965 rosso general/abstract terms 943 fare food & farming 665 trattoria life 479 fiore measurement 464 grande time 379 primavera body 350 braccio psyche 330 vedere entertainment 319 marionetta money & commerce 269 dazio communication 268 dire public domain 260 carabiniere arts & crafts 206 arte emotion 176 evviva education 135 maestro In this work, we presented the first automated analysis of code-mixing in historical travel writings. In particular, we focus on English documents about Italy, and we compare guidebooks and travel narratives, analysing the semantic categories mostly related to code-mixing.
In the future, we plan to investigate how
codemixing phenomena relate to content types in travel
writings