Morphological generation is a task where given a lemma and a morphosyntactic description of the target form, we are asked to generate the target form. Knowing that the syntactic and semantic relations to other forms are reflected by the word form itself, we show how to exploit these relations between word forms, holistically, that is, as a whole, to derive the target form without even breaking them into morphemes. Experimental results show that by organising the lexica into analogical grids we are able to improve the accuracy of morphological generation by up to 8% in low data scenarios. Our holistic approach always performs better than a morpheme-based baseline. We also enquire possible improvements by using data augmentation for neural approaches, especially in low data scenarios. However, our system seems not to gain any advantage from having more data after some point in time.

We perform a study on the universal morphological analysis task: given a word form, generate the lemma (lemmatisation) and its corresponding morphosyntactic descriptions (MSD analysis). Experiments are carried out on the SIGMORPHON 2018 Shared Task: Morphological Reinflection Task dataset which consists of more than 100 different languages with various morphological richness under three different data size conditions: low, medium and high. We consider three main approaches: morpheme-based (eager learning), holistic (lazy learning), and neural (eager learning). Performance is evaluated on the two subtasks of lemmatisation and MSD analysis. For the lemmatisation subtask, under all three data sizes, experimental results show that the holistic approach predicted more accurate lemmata, while the morpheme-based approach produced lemmata closer to the answers when it produces the wrong answers. For the MSD analysis subtask, under all three data sizes, the holistic approach achieves higher recall, while the morpheme-based approach is more precise. However, the trade-off between precision and recall of the two systems leads to a very similar overall F1 score. On the whole, neural approaches suffer under low resource conditions, but they achieve the best performance in comparison to the other approaches when the size of the training data increases.

Example-based machine translation by analogy is an alternative approach to machine translation. Its principle is relatively simple, but the absolute number of analogies between sentences contained in the corpus is crucial for the overall quality of translation. The relative number of analogies is called the analogical density. The goal of this paper is to measure the analogical density of different aligned corpora. To this end, we extract analogies between sentences. Now, we use parse trees to represent sentences on the level of syntax. We report analogical densities for five different languages in an aligned multilingual corpus extracted from the Tatoeba resource, at the level of characters, words or parse trees.

In this paper, we define a sentence re-generation task: re-generate a sentence given a set of sentences that cover it. Due to the absence of a dataset to perform this task, we firstly build three language resources of a new type containing more than 4 million annotated sentences. They contain sentences annotated with similar sentences from the same corpus, that cover them on the level of form or meaning. We then perform the sentence re-generation task on the newly produced language resources using two approaches. The first one is a na ¨ ıve approach where we rely on a language model to reorder the covering parts. The second one is a neural approach where we treat the sentence re-generation task as a translation task from a sequence of covering parts to the respective original sentence. The performance of the systems is evaluated on the level of form and meaning according to the type of covering used to re-generate the sentence. On the level of form, experimental results show that the neural approach outperforms the baseline in edit distance with up to 40% lower scores. However, in BLEU scores, the neural approach is similar or worse than the baseline. On the level of meaning, the neural approach always performs better than the baseline with average scores of 89% BERTScore.

In this paper, we inspect the theoretical problem of counting the number of analogies between sentences contained in a text. Based on this, we measure the analogical density of the text. We focus on analogy at the sentence level, based on the level of form rather than on the level of semantics. Experiments are carried on two different corpora in six European languages known to have various levels of morphological richness. Corpora are tokenised using several tokenisation schemes: character, sub-word and word. For the sub-word tokenisation scheme, we employ two popular sub-word models: unigram language model and byte-pair-encoding. The results show that the corpus with a higher Type-Token Ratio tends to have higher analogical density. We also observe that masking the tokens based on their frequency helps to increase the analogical density. As for the tokenisation scheme, the results show that analogical density decreases from the character to word. However, this is not true when tokens are masked based on their frequencies. We find that tokenising the sentences using sub-word models and masking the least frequent tokens increase analogical density.

The goal of this study is to create statistics teaching materials that incorporate an Augmented Reality (AR) system in a smartphone application. The AR system's goal is to minimize students' misconceptions about statistics in the learning process of the content of a lecture in statistics. The teaching materials include a video animation that visualizes abstract processes. The video will be shown when the markers on the module are scanned with the camera base application on the mobile phone, it can be shown. In this work, we use the problem-based learning method. This learning model is based on the misconceptions detected in the pretest and posttest activity using the Certainty Response Index (CRI) approach. The result shows that misconception thinking of the students decrease from 41% to 5% of the total number of students. The developed teaching materials have been validated from the material and content expert aspects with the acquisition of an average percentage of 92.24%, and an average of 92.71% for the media aspect.

Morphological segmentation is useful for processing Mongolian. In this paper, we manually build a morphological segmentation data set for Mongolian. We then present a character-based encoder-decoder model with attention mechanism to perform the morphological segmentation task. We further investigate the influence of analogy features extracted from scratch and improve the performance of our model using multi languages setting. Experimental results show that our encoder-decoder model with attention mechanism provides a strong baseline for Mongolian morphological segmentation. The analogy features provide useful information to the model and improve the performance of the system. The use of multi languages data set shows the capability of our model to acquire knowledge through different languages and delivers the best result.

Analogical grids are constructed from a set of words contained in a text. They tend to look like paradigm tables. Obviously, not all word forms in a language are found in a given text or corpus. This leaves empty cells inside analogical grids. Analogy is a possible way to fill in these empty cells. However, there is an issue of creating invalid word forms by analogy. This paper proposes a method to assess the validity of newly generated word forms in such empty cells. Several features are extracted from analogical grids, empty cell, and the word form itself to classify whether newly generated word form is valid or not. We carry experiments in different languages with different morphological richness. Experimental results shows that our model is able to achieve very high accuracies on invalid samples. It also delivers satisfying performance on valid samples.

Indonesian as an agglutinating language is known for its derivative morphological richness. Word forms are constructed by combining stem and affixes. In this paper, we study the influence of surface form and morphological information in analogical grids extracted from a set of word forms with varying sizes. Each word form is represented as a feature vector. In the experiment setting, we consider three features: characters, affixes, and morphosyntactic definition. The sizes and saturation are then observed to characterize the extracted grids.

The vocabulary of a natural language processing (NLP) system is usually limited by the word forms learnt by the system in the preliminary step, for example, word forms seen in the training corpus. Thus, out-of-vocabulary (OOV) problem is an important issue in NLP. In this paper, we study the plausibility of unseen word forms generated from analogical grids on Indonesian, a language known for its richness in derivational morphology. We construct analogical grids from a list of word forms contained in an annotated Indonesian corpus. We generate new word forms by filling the empty cells in the analogical grids. We verify these generated word forms using morphological analyzer and count how many of them are valid Indonesian word forms.

We release a Python module containing several tools to build analogical grids from words contained in a corpus. The module implements several previously presented algorithms. The tools are language-independent. This permits their use with any language and any writing system. We hope that the tools will ease research in morphology by allowing researchers to automatically obtain structured representations of the vocabulary contained in corpora or linguistic data. We also release analogical grids built on the vocabularies contained in 1,000 corresponding lines of the 11 different language versions of the Europarl corpus v.3. The grids were built on N-grams of different lengths, from words to 6-grams. We hope that the use of structured parallel data will foster research in comparative linguistics.

This paper presents the system submitted by IPS-WASEDA University for CoNLL– SIGMORPHON 2018 Shared Task 1: Type level inflection. We develop a system based on a holistic approach which considers wholeword form as a unit, instead of breaking them into smaller pieces (e,g. morphemes) like the baseline systems does. We also implement an encoder-decoder model which has recently become the new standard in many natural language processing (NLP) tasks. The results show that the neural approach outperforms the baseline and our holistic approach on bigger resources settings. The use of data augmentation helps to improve the performance of the model in lower resources settings, although it still cannot beat the other systems. In the end, for the low resources setting, our holistic approach performs best in comparison to the baseline and the neural approach (even with data augmentation).

Morphological inflection is the task of generating previously unseen words from morphological features. A common approach, the morpheme-based approach, decomposes words into smaller units, such as morphemes or affixes, learned in advance. This paper proposes a different approach. It shows that breaking words into pieces is not necessarily the best option. The proposed approach is holistic and treats whole word forms as basic units in its description of morphological variations among word forms. The approach generates inflected forms by solving analogical equations between whole word forms; morphological features can be used as constraints. Experiment results on the 52 languages of SIGMORPHON 2017 Shared Task show that the proposed approach performs as good as the morpheme-based approach, even slightly better on average. This demonstrates the absence of necessity of explicitly learning how to decompose words.

. Analogical grids aim to capture the organization of the lexicon of a language. We conduct experiments on analogical grids extracted in four different languages with different morphological richness. We study the saturation of analogical grids against their size. We observe that the logarithm of the saturation of an analogical grid is linear in the logarithm of its size. More surprisingly, the coefficients of this log-log linear relation are extremely close across all four languages, even when the size or the genre of the corpus vary

We propose a pipeline to explain, on the level of form, the unseen words contained in an Indonesian test set, by using analogical clusters. Analogical clusters are extracted from a training set by relying on formal relations between words. The unseen words which can be explained on the level of form are then verified on two other representation levels: morphological and semantic. In our experiments on the BPPT corpus, 98 % of unseen words were explained on the level form, out of which 58 % could also be explained on the two levels of morphological and semantic representations.

We address the problem of predicting unseen words by relying on the organization of the vocabulary of a language as exhibited by paradigm tables. We present a pipeline to automatically produce paradigm tables from all the words contained in a text. We measure how many unseen words from an unseen test text can be predicted using the paradigm tables obtained from a training text. Experiments are carried out in several languages to compare the morphological richness of languages, and also the richness of the vocabulary of di↵erent authors.

This paper describes work on a part-of-speech tagger for the Indonesian language by employing a rule-based approach. The system tokenizes documents while also considering multi-word expressions and recognizes named entities. It then applies tags to every token, starting from closed-class words to open-class words and disambiguates the tags based on a set of manually defined rules. The system currently obtains an accuracy of 79% on a manually tagged corpus of roughly 250.000 tokens.

We describe our work on designing a linguistically principled part of speech (POS) tagset for the Indonesian language. The process involves a detailed study and analysis of existing tagsets and the manual tagging of an Indonesian corpus. The results of this work are an Indonesian POS tagset consisting of 23 tags and an Indonesian corpus of over 250.000 lexical tokens that have been manually tagged using this tagset.

Pemuisi is a poetry generation system that generates topical poems in Indonesian using a constraint satisfaction approach. It scans popular news websites for articles and extracts relevant keywords that are combined with various language resources such as templates and other slot fillers into lines of poetry. It then composes poems from these lines by satisfying a set of given constraints. A Turing Test-style evaluation and a detailed evaluation of three different configurations of the system was conducted through an online questionnaire with 180 respondents. The results showed that under the best scenario, 57% of the respondents thought that the generated poems were authored by humans, and that poems generated using the full set of constraints consistently measured better on all aspects than those generated using the other two configurations. The system is now available online as a web application.