Mixed Script Identification Using Automated DNN Hyperparameter Optimization

Mixed script identification is a hindrance for automated natural language processing systems. Mixing cursive scripts of different languages is a challenge because NLP methods like POS tagging and word sense disambiguation suffer from noisy text. This study tackles the challenge of mixed script identification for mixed-code dataset consisting of Roman Urdu, Hindi, Saraiki, Bengali, and English. The language identification model is trained using word vectorization and RNN variants. Moreover, through experimental investigation, different architectures are optimized for the task associated with Long Short-Term Memory (LSTM), Bidirectional LSTM, Gated Recurrent Unit (GRU), and Bidirectional Gated Recurrent Unit (Bi-GRU). Experimentation achieved the highest accuracy of 90.17 for Bi-GRU, applying learned word class features along with embedding with GloVe. Moreover, this study addresses the issues related to multilingual environments, such as Roman words merged with English characters, generative spellings, and phonetic typing.

The TRANSFER OF CODES AND MIXED CODES IN INDONESIAN MOLEBRITIES INSTAGRAM

This study tries to analyze the code and code mix in Indonesian celebrity Instagram accounts. The research method used is descriptive. Is this research. Then, the source of this research data is uploading on celebrity Instagram accounts from May to June 2020. Meanwhile, the data in this study are the status information uploaded by Indonesian celebrities. Furthermore, collecting data in this study was carried out by making screenshots of uploaded photos or videos accompanied by information and data sourced from Indonesian celebrity Instagram accounts. Then, all the collected text will be categorized by the type of code switching and mixed code. Based on the analysis of research results can be seen as follows. (1) code switching found in the Indonesian celebrity Instagram account caption upload caption over internal code and external code transfer, (2) mixed code found in the Indonesian celebrity idiom Instagram Instagram account caption, and clauses.

JENIS DAN POLA CAMPUR KODE DALAM PERCAKAPAN GURU DAN SISWA PADA WHATSAPP GRUP

This research was conducted to determine the code mixing patterns contained in the conversation between teachers and students on WhatsApp group class XII SMK Muhammadiyah 2 Andong. This research belongs to a qualitative descriptive study. The data used in the management of this research is in the form of writing, which is in the form of digital communication between teachers and students on the Whatsapp group media. The data sources of this research came from teachers and students of class XII SMK Muhammadiyah 2 Andong. The data were collected using listening techniques, tapping techniques, listening techniques without proficient involvement, and note taking techniques. The data analysis technique used is the BUL technique as the basic technique and the non-parafrasal form changing technique as a follow-up technique. The results of the analysis show that there are 3 type of code mixing with 9 patterns of mixing code in the speech. The code mix is divided into 2 patterns, namely Java-Indonesia-Java and Java-Indonesia-Java-Indonesia-Java-Indonesia. Mix code out is divided into 4 patterns, including: 1) Indonesian-English-Indonesian; 2) Javanese-English-Javanese; 3) English-Indonesian-English-Indonesian-English-Indonesian; 4) English-Javanese-English-Javanese. Meanwhile, mixed code mix was found in 3 patterns, including: 1) Indonesian-Javanese-English; 2) Indonesian-English-Javanese; and 3) Indonesian-English-Indonesian-Javanese.

An Algorithm for the Detection of Hidden Propaganda in Mixed-Code Text over the Internet

Internet-based communication systems have become an increasing tool for spreading misinformation and propaganda. Even though there exist mechanisms that are able to track unwarranted information and messages, users made up different ways to avoid their scrutiny and detection. An example is represented by the mixed-code language, that is text written in an unconventional form by combining different languages, symbols, scripts and shapes. It aims to make more difficult the detection of specific content, due to its custom and ever changing appearance, by using special characters to substitute for alphabet letters. Indeed, such substitute combinations of symbols, which tries to resemble the shape of the intended alphabet’s letter, makes it still intuitively readable to humans, however nonsensical to machines. In this context, the paper explores the possibility of identifying propaganda in such mixed-code texts over the Internet, centred on a machine learning based approach. In particular, an algorithm in combination with a deep learning models for character identification is proposed in order to detect and analyse whether an element contains propaganda related content. The overall approach is presented, the results gathered from its experimentation are discussed and the achieved performances are compared with the related works.

CAMPUR KODE PERULANGAN KATA DALAM NOVEL RADIKUS MAKAN KAKUS KARYA: RADITYA DIKA (KAJIAN SOSIOLINGUISTIK)

This study aims to obtain a description of the type and form of mixed reduplication word code in the novel RadikusMakankakus, by RadityaDika. This research data comes from RadikusMakankakus novel, RadityaDika. Data collection in this study using the method refer to the basic techniques of tangible techniques tapping techniques and techniques as advanced techniques. Based on the results of the research, that in the novel RadikusMakankakus, found the type of mixed code into (inner code maxing) and the type of mixed code out (outer code maxing). Type of mixed code into (inner code maxing) found, among others, is mixed code into (inner code maxing) which tangible repetition of Indonesian dialect Jakarta, Javanese words, Jakarta dialect. In addition, also found type of mixed code out (outer code maxing). The outer code maxing types found in the novel include mixing outward code in the form of looping of English words.

ALIH KODE DAN CAMPUR KODE PADA NOVEL BADAI MATAHARI ANDALUSIA KARYA HARY EL-PARSIA (Transfer code and mix code in Novels Badai Matahari Andalusia Karya Hary El-Parsia)

The language produced by speakers is a reflection of a culture that it adheres to because, incommunication, it certainly has a close relationship with the other person so that the language spoken can bewell understood. A discussion can take place if the style used can be followed so that the message delivered canbe received. This study aims to analyze code-switching and code-mixing in novels Badai Matahari CreationHary El-Parsia. The method in this research is descriptive qualitative carried out by not prioritizing the numbersbut prioritizes the depth of appreciation of the interaction between the concepts being studied imply. The type ofresearch in the literature is. The source and type of data of this research are the data written on novel BadaiMatahari Andalusia. The results of the study showed that in the Solar Storm novel, there were seven quotes fromthe Code and nine quotes from the Code. Code Transfer and Mix The most dominant code is the Transfer ofCode by mixing Indonesian, English, Arabic. In other words, the author Hary El-Parsia in his work uses theTransfer of Code and Mixed Code as the hallmark of his work.Keywords: Code Transfer, Code Combination. Novel. Badai Matahari Andalusia

ALIH KODE DAN CAMPUR KODE PADA NOVEL BADAI MATAHARI ANDALUSIA KARYA HARY EL-PARSIA (Transfer code and mix code in Novels Badai Matahari Andalusia Karya Hary El-Parsia)

The language produced by speakers is a reflection of a culture that it adheres to because, incommunication, it certainly has a close relationship with the other person so that the language spoken can bewell understood. A discussion can take place if the style used can be followed so that the message delivered canbe received. This study aims to analyze code-switching and code-mixing in novels Badai Matahari CreationHary El-Parsia. The method in this research is descriptive qualitative carried out by not prioritizing the numbersbut prioritizes the depth of appreciation of the interaction between the concepts being studied imply. The type ofresearch in the literature is. The source and type of data of this research are the data written on novel BadaiMatahari Andalusia. The results of the study showed that in the Solar Storm novel, there were seven quotes fromthe Code and nine quotes from the Code. Code Transfer and Mix The most dominant code is the Transfer ofCode by mixing Indonesian, English, Arabic. In other words, the author Hary El-Parsia in his work uses theTransfer of Code and Mixed Code as the hallmark of his work.Keywords: Code Transfer, Code Combination. Novel. Badai Matahari Andalusia

Bentuk Campur Kode Dalam Buku Resep Mindy Cake & Cookies Karya Mindy Mot

Indonesians, in general, are bilinguals. This is because Indonesians have both mother tongue and second language which is used nationally. A person who can speak two languages or more in both formal and informal is called a bilingual. Even though each language has its respective roles, if it is used in one speech or writing it will cause codemixing. Code mixing is the use of two or more languages in one speech. In this study, the researcher analyzed the Mindy Cake & Cookies recipe book by Mindy Mot. The researcher chose Mindy Cake & Cookies recipe book by Mindy Mot because this book is very popular among teenagers, foodies and various circles. In 2019 the Mindy Mot recipe became one of the trending topics on Instagram and YouTube. This recipe book uses Indonesian and a foreign language which causes code-mixing. This study aims to understand and analyze the forms of code-mixing, the types of code-mixing, and the factors causing codemixing in the Mindy Cake & Cookies recipe book by Mindy Mot. The data collection technique in this research was a literature study. The method used in the mixed code research in Mindy Cake & Cookies recipe book by Mindy Mot is a qualitative descriptive method. The object of research in this study is the Mindy Cake & Cookies recipe book by Mindy Mot. The results of the analysis reveal that there are some forms of code-mixing in the Mindy Cake and Cookies by Mindy Mot, such as the form of words and phrases. The code-mixing form also comes in the form of words. There are 15 words out of 26 words which are considered as code-mixing. They are baking, chiffon, muffin, cake, mixer, spatula, stainless, fresh, frosting, meringue, filling, raspberry, whisk, crumble, freezer. There are also found 10 phrases out of 17 phrases that are included in the code-mixing. Those are baking powder, baking soda, overmix, buttercream, whipping cream, buttercream baileys, baileys buttercream, cake emulsifier, cream cheese, brown sugar. Keywords: Code-Mixing, Recipe Book, Cake, Cookies

FENOMENA BILINGUALISME (CAMPUR-ALIH KODE) BAHASA DALAM PERCAKAPAN SANTRI DI PONDOK PESANTREN AL MANSHUR POPONGAN KLATEN

This study aims to discuss one of the phenomena of bilingualism (code switching) and interference as well as the factors causing the phenomenon of language bilingualism in santri conversations in Islamic boarding schools. The forms of bilingualism and interference include Javanese, Indonesian and Arabic. The approach in this research is descriptive qualitative which describes the state of the object of research based on the facts that appear as they are. The subjects of this study were the students of Al Manshur Popongan Klaten. Data collection techniques in the form of observation, interviews and documentation. The analysis technique during the field uses the model of Miles and Huberman, namely data collection, data reduction, data presentation and conclusion or verification. The results of the study note that there are 55 conversational data in the Al Manshur Popongan Islamic boarding school classified as follows; internal code 7 data transfer, external code 10 data transfer. Mix codes in linguistic elements that occur namely mixed code insertion of 12 data tangible elements, mixed code 5 data tangible codes, mixed code insertion elements in clause 7 data and mixed code insertion elements in the 5 word repetition element. Factors causing code switching and code mixing are the presence of speakers, interlocutors, prestige, changes in topic of conversation, identification of social roles, the desire to explain, and the habit of speakers.