Applied AI letters最新文献

Amharic is one of the under-resourced languages, making news headline generation particularly challenging due to the scarcity of high-quality linguistic datasets necessary for training effective natural language processing models. In this study, we fine-tuned the small check point of the T5v1.1 model (t5-small) to perform Amharic news headline generation with an Amharic dataset that is comprised of over 70k news articles along with their headline. Fine-tuning the model involves dataset collection from Amharic news websites, text cleaning, news article size optimization using the TF-IDF algorithm, and tokenization. In addition, a tokenizer model is developed using the byte pair encoding (BPE) algorithm prior to feeding the dataset for feature extraction and summarization. Metrics including Rouge-L, BLEU, and Meteor were used to evaluate the performance of the model and a score of 0.5, 0.24, and 0.71, respectively, was achieved on the test partition of the dataset that contains 7230 instances. The results were good relative to result of the t5 model without fine-tuning, which are 0.1, 0.03, and 0.14, respectively. A postprocessing technique using a rule-based approach was used for further improving summaries generated by the model. The addition of the postprocessing helped the system to achieve Rouge-L, BLEU, and Meteor scores of 0.72, 0.52, and 0.81, respectively. The result value is relatively better than the result achieved by the nonfine-tuned T5v1.1 model and the result of previous studies report on abstractive-based text summarization for Amharic language, which had a 0.27 Rouge-L score. This contributes a valuable insight for practical application and further improvement of the model in the future by increasing the article length, using more training data, using machine learning–based adaptive postprocessing techniques, and fine-tuning other available pretrained models for text summarization.

Malware growth has accelerated due to the widespread use of Android applications. Android smartphone attacks have increased due to the widespread use of these devices. While deep learning models offer high efficiency and accuracy, training them on large and complex datasets is computationally expensive. Hence, a method that effectively detects new malware variants at a low computational cost is required. A transfer learning method to detect Android malware is proposed in this research. Because of transferring known features from a source model that has been trained to a target model, the transfer learning approach reduces the need for new training data and minimizes the need for huge amounts of computational power. We performed many experiments on 1.2 million Android application samples for performance evaluation. In addition, we evaluated how well our framework performed in comparison with traditional deep learning and standard machine learning models. In comparison with state-of-the-art Android malware detection methods, the proposed framework offers improved classification accuracy of 98.87%, a precision of 99.55%, recall of 97.30%, F1-measure of 99.42%, and a quicker detection rate of 5.14 ms using the transfer learning strategy.

Africa has over 2000 languages; however, those languages are not well represented in the existing natural language processing ecosystem. African languages lack essential digital resources to effectively engage in advancing language technologies. There is a need to generate high-quality natural language processing resources for low-resourced African languages. Obtaining high-quality speech and text data is expensive and tedious because it can involve manual sourcing and verification of data sources. This paper discusses the process taken to curate and annotate text and speech datasets for five East African languages: Luganda, Runyankore-Rukiga, Acholi, Lumasaba, and Swahili. We also present results obtained from baseline models for machine translation, topic modeling and classification, sentiment classification, and automatic speech recognition tasks. Finally, we discuss the experiences, challenges, and lessons learned in creating the text and speech datasets.

We present Clipboard-to-SMILES Converter (C2SC), a macOS application that directly converts molecular structures from the clipboard. The app focuses on seamlessly converting screenshots of molecules into a desired molecular representation. It supports a wide range of molecular representations, such as SMILES, SELFIES, InChI's, IUPAC names, RDKit Mol's, and CAS numbers, allowing effortless conversion between these formats within the clipboard. C2SC automatically saves converted molecules to a local history file and displays the last 10 entries for quick access. Additionally, it incorporates several SMILES operations, including canonicalization, augmentation, as well price-searching molecules on chemical vendors for the cost-effective purchasing option. Beyond the one-click conversion from clipboard to molecular structures, the app offers continuous monitoring of the clipboard which automatically converts any supported representations or images detected into SMILES. The convenient interface, directly in the status bar, as well as availability as macOS application, makes C2SC useful for a broad community not requiring any programming expertise. Most conversions are performed locally, notably the image-to-SMILES conversion, with internet access only necessary for specific tasks like price lookups. In summary, C2SC provides a user-friendly and efficient solution for converting molecular structures directly from the clipboard, offering seamless conversions between comprehensive chemical representations and can be directly downloaded from https://github.com/O-Schilter/Clipboard-to-SMILES-Converter.

Users of e-commerce websites review different aspects of a product in the comment section. In this research, an approach is proposed for opinion aspect extraction and recognition in selling systems. We have used the users' opinions from the Digikala website (www.Digikala.com), which is an Iranian e-commerce company. In this research, a language-independent framework is proposed that is adjustable to other languages. In this regard, after necessary text processing and preparation steps, the existence of an aspect in an opinion is determined using deep learning algorithms. The proposed model combines Convolutional Neural Network (CNN) and long-short-term memory (LSTM) deep learning approaches. CNN is one of the best algorithms for extracting latent features from data. On the other hand, LSTM can detect latent temporal relationships among different words in a text due to its memory ability and attention model. The approach is evaluated on six classes of opinion aspects. Based on the experiments, the proposed model's accuracy, precision, and recall are 70%, 60%, and 85%, respectively. The proposed model was compared in terms of the above criteria with CNN, Naive Bayes, and SVM algorithms and showed satisfying performance.

The ease with which mobile money is used to facilitate cross-border payments presents a global threat to law enforcement in the fight against money laundering and terrorist financing. This paper aims to utilize machine learning classifiers to predict transactions flagged as a fraud in mobile money transfers. The data for this study were obtained from real-time transactions that simulate a well-known mobile transfer fraud scheme. Logistic regression is used as the baseline model and is compared with ensemble and gradient descent models. The results indicate that the logistic regression model still showed reasonable performance while not performing as well as the other models. Among all the measures, the random forest classifier exhibited outstanding performance. The amount of money transferred emerged as the top feature for predicting money laundering transactions in mobile money transfers. These findings suggest that further research is needed to enhance the logistic regression model, and the random forest classifier should be explored as a potential tool for law enforcement and financial institutions to detect money laundering activities in mobile money transfers.

Patents show how technology evolves in most scientific fields over time. The best way to use this valuable knowledge base is to use efficient and effective information retrieval and searches for related prior art. Patent classification, that is, assigning a patent to one or more predefined categories, is a fundamental step towards synthesizing the information content of an invention. To this end, architectures based on Transformers, especially those derived from the BERT family have already been proposed in the literature and they have shown remarkable results by setting a new state-of-the-art performance for the classification task. Here, we study how domain adaptation can push the performance boundaries in patent classification by rigorously evaluating and implementing a collection of recent transfer learning techniques, for example, domain-adaptive pretraining and adapters. Our analysis shows how leveraging these advancements enables the development of state-of-the-art models with increased precision, recall, and F1-score. We base our evaluation on both standard patent classification datasets derived from patent offices-defined code hierarchies and more practical real-world use-case scenarios containing labels from the agrochemical industrial domain. The application of these domain adapted techniques to patent classification in a multilingual setting is also examined and evaluated.