A New Translation Strategy Based-on Case-Based Reasoning in Machine Translation System

Automatically assessing the quality of machine translation is an important yet challenging task for machine translation research. Translation quality assessment is understood as predicting translation quality without reference to the source text. Translation quality depends on the specific machine translation system and often requires post-editing. Manual editing is a long and expensive process. Since the need to quickly determine the quality of translation increases, its automation is required. In this paper, we propose a quality assessment method based on ensemble supervised machine learning methods. The bilingual corpus WMT 2019 for the EnglishRussian language pair was used as data. The text data volume is 17089 sentences, 85% of the data was used for training, and 15% for testing the model. Linguistic functions extracted from the text in the source and target languages were used as features for training the system, since it is these characteristics that can most accurately characterize the translation in terms of quality. The following tools were used for feature extraction: a free language modeling tool based on SRILM and a Stanford POS Tagger parts of speech tagger. Before training the system, the text was preprocessed. The model was trained using three regression methods: Bagging, Extra Tree, and Random Forest. The algorithms were implemented in the Python programming language using the Scikit learn library. The parameters of the random forest method have been optimized using a grid search. The performance of the model was assessed by the mean absolute error MAE and the root mean square error RMSE, as well as by the Pearsоn coefficient, which determines the correlation with human judgment. Testing was carried out using three machine translation systems: Google and Bing neural systems, Mouses statistical machine translation systems based on phrases and based on syntax. Based on the results of the work, the method of additional trees showed itself best. In addition, for all categories of indicators under consideration, the best results are achieved using the Google machine translation system. The developed method showed good results close to human judgment. The system can be used for further research in the task of assessing the quality of translation.

This paper describes a method for the automatic inference of structural transfer rules to be used in a shallow-transfer machine translation (MT) system from small parallel corpora. The structural transfer rules are based on alignment templates, like those used in statistical MT. Alignment templates are extracted from sentence-aligned parallel corpora and extended with a set of restrictions which are derived from the bilingual dictionary of the MT system and control their application as transfer rules. The experiments conducted using three different language pairs in the free/open-source MT platform Apertium show that translation quality is improved as compared to word-for-word translation (when no transfer rules are used), and that the resulting translation quality is close to that obtained using hand-coded transfer rules. The method we present is entirely unsupervised and benefits from information in the rest of modules of the MT system in which the inferred rules are applied.

Corpus-based machine translation (MT) has been the main approach to developing and implementing MT systems in both academia and the industry over the last three decades. In this field, the type and size of the corpus used for training MT engines have presented problems for both statistical MT (SMT) systems as well as neural MT (NMT) systems, being the two dominant corpusbased approaches. Moreover, language pairs such as Turkish-English have been understudied within this framework. This article aims to evaluate the translation quality in Turkish-to-English custom MT systems that have been trained on different corpus sizes and types. Two NMT engines and two SMT engines were trained on the KantanMT platform using two different training corpus types with either only domain-specific cardiology corpus or this corpus plus a mixed-domain corpus. The study conducted both automatic evaluations with metrics including BLEU, F-Measure and TER, as well as a comprehensive human evaluation with metrics including fluency, A/B test, and adequacy. Lastly, the study realized a separate, subjective terminology evaluation in order to investigate how differently MT systems handle terminology, as this is a crucial aspect for specific-domain text types such as cardiology. While the automatic evaluation results suggest the SMT engines to perform better than NMT engines, all human evaluators rated the mixed-domain NMT engine as the highest performing one. However, the terminology evaluation task demonstrated SMT to still be able to perform better and to commit less terminology errors, despite the industry and academia shifting toward NMT engines.

Query expansion is a classical topic in the field of information retrieval, which is proposed to bridge the gap between searchers' information intents and their queries. Previous researches usually expand queries based on document collections, or some external resources such as WordNet and Wikipedia [1, 2, 3, 4, 5]. However, it seems that independently using one of these resources has some defects, document collections lack semantic information of words, while WordNet and Wikipedia may not include domain-specific knowledge in certain document collection. Our work aims to combine these two kinds of resources to establish an expansion model which represents not only domain-specific information but also semantic information. In our preliminary experiments, we construct a two-layer word graph and use Random-Walk algorithm to calculate the weights of each term in pseudo-relevance feedback documents, then select the highest weighted term to expand original query. The first layer of the word graph contains terms in related documents, while the second layer contains semantic senses corresponding to these terms. These terms and semantic senses are treated as vertices of the graph and connected with each other by all possible relationships, such as mutual information and semantic similarities. We utilized mutual information, semantic similarity and uniform distribution as the weight of term-term relation, sense-sense relation and word-sense relation respectively. Though these experiments show that our expansion outperform original queries, we are troubled with some difficult problems. Given the framework of semantic graph model, we need more effort to find out an optimal graph to represent the relationships between terms and their semantic senses. We utilized a two-layer graph model in our preliminary research, where terms from different documents are treated equally. Maybe we can introduce the document as a third layer in future work, where we can differ the same terms in different documents according to document relevance and context. Then we need appropriately represent initial weights of this words, senses and relationships. Various measures for weights of terms and term relations have been proved effective in other information retrieval tasks, such as TFIDF, mutual information (MI), but there is little research on weights for semantic senses and their relations. For polysemous words, we add all of their semantic senses to the graph and assume that these senses are uniformly distributed. Actually, it is not precise for a word in a special document and query. As we know, a polysemous word may have only one or two senses in a document, and they are not uniformly distributed. Give a word, what we should do is to determine its word senses in a relevant document and estimate the distribution of these senses. Word sense disambiguation may help us in this problem. Then, there are many methods to compute word similarity according to WordNet, which we use to represent the weights of relationships between word senses. Varelas et al implemented some popular methods to compute semantic similarity by mapping terms to an ontology and examining their relationships in that ontology [4]. We also need to know which algorithm for semantic similarity is most suitable for our model. Additional, WordNet is suitable to calculate word similarity but not suitable to measure word relevance. The inner hyperlinks of Wikipedia could help us to calculate word relevance. We wish to find an effective way to combine the similarity measure from WordNet and relevance measure from Wikipedia, which may completely reflect word relationships.

Dynamic programming (DP) is a fundamental tool used to obtain exact, optimal solutions for many combinatorial optimization problems. Among these problems, important ones including the knapsack problems and the computation of edit distances between string pairs can be solved with a kind of DP that corresponds to solving the shortest path problem on a directed acyclic graph (DAG). These problems can be solved efficiently with DP, however, in practical situations, we want to solve the customized problems made by adding logical constraints to the original problems. Developing an algorithm specifically for each combination of a problem and a constraint set is unrealistic. The proposed method, BDD-Constrained Search (BCS), exploits a Binary Decision Diagram (BDD) that represents the logical constraints in combination with the DAG that represents the problem. The BCS runs DP on the DAG while using the BDD to check the equivalence and the validity of intermediate solutions to efficiently solve the problem. The important feature of BCS is that it can be applied to problems with various types of logical constraints in a unified way once we represent the constraints as a BDD. We give a theoretical analysis on the time complexity of BCS and also conduct experiments to compare its performance to that of a state-of-the-art integer linear programming solver.

A workflow trace describes provenance information of a particular workflow execution. Understanding workflow traces and their similarity have many applications in both scientific research and business world. Given workflow traces generated by heterogeneous systems with difference granularities, it is a challenge for users to understand their similarities. In this work, we investigate workflow traces’ granularity problem and their similarity method. Algorithms are developed to transform a trace into its multi-granularity forms assisting by a workflow trace ontology. A novel generic semantic similarity algorithm is proposed that not only considers the structural similarity but also the semantics coverage embedded in traces during transformation. Furthermore, theoretical analysis is presented to compute the maximum semantic similarity. Our approach enables that two workflow traces can be compared with any granularity. The experiment using real world workflow traces demonstrates the effectiveness of the proposed methods.

The modern world is undergoing an active transformation and development of trade, cultural, economic and political ties between states. Practically each of these areas is regulated by a multitude of laws and regulations, which is reflected in various types of legal documents and texts. Given that participants in international relations represent different linguistic and cultural backgrounds, the need for high-quality legal translation is becoming more urgent. In the modern world, machine translation is becoming increasingly relevant, including in the legal field, taking into account the complexities of judicial language, cultural differences and the peculiarities of legal systems of states. The purpose of this article is to determine the effectiveness of various machine translation systems when working with legal texts. The focus of this work is on two types of legal texts belonging to different functional styles, which allows the authors to trace the dependence of the adequacy and equivalence of machine translation on the communicative situation. A comparative analysis of the quality of translation of two types of professional texts performed by humans and machine translation systems is carried out. The methodological basis of the study was the classification method, descriptive and comparative methods. The research results will help professional translators or ordinary users choose the highest quality machine translation service, identify and eliminate typological errors that occur when working with texts of a certain subject at the stages of pre-editing and post-editing.

The article presents a comparative analysis of various online translators in the machine translation system. The author of the study focused on finding out the effectiveness and accuracy of the translation, identifying the strengths and weaknesses of the three most common online translators: Google Translate, DeepL and Microsoft Translator. The article begins with an overview of current trends in the field of machine translation and the need for online translations based on this context. The research methodology is selected in the article, in particular, the selection of individual online translators for comparison and the criteria for evaluating their work. The researcher analyzes the quality of translations for various types of texts, including common phrases, technical terms, complex sentences, etc. Attention is also paid to different language pairs and ensuring the availability of rarer languages in electronic translation. The article reveals such aspects as accuracy of translation, speed of work, recognition of context and idiomatic expressions, availability of additional functions and capabilities that ensure better translation quality. The author thoroughly highlights the strengths and weaknesses of each system, provides clear information about the use of integrations, available functions and additional capabilities of each translator. In the final part of the article, the author provides conclusions and recommendations regarding the use of online translators in the machine translation system. The research highlights that each electronic translator has its own advantages and limitations, and the choice depends on the specific needs of the user. In addition, the author emphasizes the need for constant updating and improvement of online translators to increase their accuracy and efficiency. This comparative analysis will help simplify the selection of the optimal translation software for users who require high-quality machine translation for their professional activities or everyday needs.

To make ECommerce information searching across Internet more efficient, ECommerce information searching becomes more and more important. In this paper, ECommerce information model (EIM) and a novel EIM-based semantic similarity algorithm are presented. This semantic similarity algorithm takes advantage of ECommerce-based information content and edge-based distance in calculating conceptual similarity. According to EIM, a semantic eigenvector, which consists of the semantic similarity values of a given document, is used to represent the semantic content of the document. The semantic eigenvectors and EIM-based similarity function can be applied to ECommerce information retrieval. Experimental results show that the performance of the proposed method is much improved when compared with that of the traditional information retrieval techniques

Machine Translation (MT) which has been championed as an effective technology for knowledge transfer from English to languages with less digital content. An example of such efforts is the automatic translation of English Wikipedia to languages with smaller collections, such as Arabic. However, MT quality is still far from ideal for many of the languages and text genres. While translating a document, many sentences are poorly translated which can provide an incorrect text, and confuse the reader. Moreover, some of these sentences are not as informative and could be summarized to make a more cohesive document. Thus, for tasks in which complete translation is not mandatory, MT can be effective if the system can provide an informative subset of the content with higher translation quality. For this scenario, text summarization can provide effective support for MT by keeping only the most important and informative parts of a given document to translate. In this work, we demonstrate a framework of MT and text summarization which replaces the baseline translation with a proper summary that has higher translation quality than the full translation. For this, we combine the state of the art English summarization system and a novel framework for prediction of MT quality without references. Our framework is composed of the following major components: (a) a standard machine translation system, (b) a reference-free MT quality estimation system, (c) an MT-aware summarization system, and (d) an English-Arabic sentence matcher. More specifically, our English-Arabic system reads in an English document along with its baseline Arabic translation and outputs, as a summary, a subset of the Arabic sentences based on their informativeness and their translation quality. We demonstrate the utility of our system by evaluating it with respect to its translation and summarization quality and demonstrate that we can balance between improving MT quality and maintaining a decent summarization quality. For summarization, we conduct both reference-based and reference-free evaluations and observe a performance in the range of the state of the art system. Moreover, the translation quality of the summaries shows an important improvement against the baseline translation of the entire documents. This MT-aware summarization approach can be applied to translation of texts such as Wikipedia articles. For such domain-rich articles, there is a large variation of translation quality across different sections. An intelligent reduction of the translation tasks results in improved final outcome. Finally, the framework is mostly language independent and can be easily customized for different target languages and domains.

Most commonly, case-based reasoning is applied in domains where attribute value representations of cases are sufficient to represent the features relevant to support classification, diagnosis or design tasks. Distance functions like the Hamming-distance or their transformation into similarity functions are applied to retrieve past cases to be used to generate the solution of an actual problem. Often, domain knowledge is available to adapt past solutions to new problems or to evaluate solutions. However, there are domains like architectural design or law in which structural case representations and corresponding structural similarity functions are needed. Often, the acquisition of adaptation knowledge seems to be impossible or rather requires an effort that is not manageable for fielded applications. Despite of this, humans use cases as the main source to generate adapted solutions. How to achieve this computationally? This paper presents a general approach to structural similarity assessment and adaptation. The approach allows to explore structural case representations and limited domain knowledge to support design tasks. It is exemplarily instantiated in three modules of the design assistant FABEL-Idea that generates adapted design solutions on the basis of prior CAD layouts.

Localized machine translation technologies have been developed to address region specific linguistic needs, yet they remain less widely adopted than dominant global systems. Despite their potential, limited evidence exists regarding their translation quality, particularly in complex communicative contexts. This study addresses this gap by evaluating the performance of three domestic machine translation systems and examining their strengths, limitations, and overall reliability in English to Persian translation. A set of eighty test statements representing narrative, descriptive, argumentative, and idiomatic forms was used to assess linguistic accuracy across grammatical, semantic, and pragmatic dimensions. The findings indicate that although the systems generated broadly intelligible translations, they showed recurrent weaknesses in subject verb agreement, tense consistency, and especially pragmatic rendering, where idiomatic expressions posed substantial challenges. One system demonstrated comparatively stronger syntactic and semantic performance, yet none provided consistently adequate output without the need for substantial post editing. For comparative purposes, translations edited through a large-scale language model displayed improved grammatical correctness and reduced error rates, although occasional literal renderings required minor adjustments. The overall analysis suggests that current local technologies do not yet deliver dependable high quality translations independently and that significant technological refinement is required to enhance their usability for professional or high stakes applications. Keywords: Artificial intelligence; language technology; machine translation; translation evaluation; translation quality.

Constructing a corpus of parallel sentence pairs is an important work in building a Statistical Machine Translation system. It impacts deeply how the quality of a Statistical Machine Translation could achieve. The more parallel sentence pairs we use to train the system, the better translation's quality it is. Nowadays, comparable non-parallel corpora become important resources to alleviate scarcity of parallel corpora. The problem here is how to extract parallel sentence pairs automatically but accurately from comparable non-parallel corpora, which are usually very noisy. This paper presents how we can apply the reinforcement-learning scheme with our new proposed algorithm for detecting parallel sentence pairs. We specify that from an initial set of parallel sentences in a domain, the proposed model can extract a large number of new parallel sentence pairs from non-parallel corpora resources in different domains, concurrently increasing the system's translation ability gradually.

The integration of machine translation (MT) and artificial intelligence (AI) has significantly enhanced the precision of MT systems producing translations that rival the quality of skilled human translators. This innovation has broadened the applicability of AI-enabled MT systems attracting users for a variety of purposes. This paper explores the following areas: the application of AI-enabled MT, specifically Google Translate (GT) in English as a Foreign Language (EFL) classroom. Empirical findings and recent research indicate that students are increasingly reliant on MT particularly in higher educational settings where English is not their first language. The purpose of this research is to provide insight on the practices, beliefs and goals of EFL learners who use GT in the classroom. Data was gathered from 234 university students through a questionnaire along with pre- and post-tests to compare the quality of writing drafts created with and without the aid of GT. The findings are consistent with those of previous research conducted worldwide highlighting the immense potential of AI-enabled MT in not only boosting students' learning experiences but also cultivating more independent learners. The study also indicates that students have positive attitudes towards GT and employ diverse search strategies to address various language-related challenges.

We present a method to incorporate nonlinear shape prior constraints into segmenting different anatomical structures in medical images. Kernel space density estimation (KSDE) is used to derive the nonlinear shape statistics and enable building a single model for a class of objects with nonlinearly varying shapes. The object contour is coerced by image-based energy into the correct shape sub-distribution (e.g., left or right lung), without the need for model selection. In contrast to an earlier algorithm that uses a local gradient-descent search (susceptible to local minima), we propose an algorithm that iterates between dynamic programming (DP) and shape regularization. DP is capable of finding an optimal contour in the search space that maximizes a cost function related to the difference between the interior and exterior of the object. To enforce the nonlinear shape prior, we propose two shape regularization methods, global and local regularization. Global regularization is applied after each DP search to move the entire shape vector in the shape space in a gradient descent fashion to the position of probable shapes learned from training. The regularized shape is used as the starting shape for the next iteration. Local regularization is accomplished through modifying the search space of the DP. The modified search space only allows a certain amount of deformation of the local shape from the starting shape. Both regularization methods ensure the consistency between the resulted shape with the training shapes, while still preserving DP’s ability to search over a large range and avoid local minima. Our algorithm was applied to two different segmentation tasks for radiographic images: lung field and clavicle segmentation. Both applications have shown that our method is effective and versatile in segmenting various anatomical structures under prior shape constraints; and it is robust to noise and local minima caused by clutter (e.g., blood vessels) and other similar structures (e.g., ribs). We believe that the proposed algorithm represents a major step in the paradigm shift to object segmentation under nonlinear shape constraints.