Anna Wróblewska - Academia.edu (original) (raw)

Papers by Anna Wróblewska

Lecture Notes in Computer Science, 2022

Cornell University - arXiv, Jun 9, 2022

Creating a meaningful representation by fusing single modalities (e.g., text, images, or audio) i... more Creating a meaningful representation by fusing single modalities (e.g., text, images, or audio) is the core concept of multimodal learning. Although several techniques for building multimodal representations have been proven successful, they have not been compared yet. Therefore it has been ambiguous which technique can be expected to yield the best results in a given scenario and what factors should be considered while choosing such a technique. This paper explores the most common techniques for building multimodal data representations-the late fusion, the early fusion, and the sketch, and compares them in classification tasks. Experiments are conducted on three datasets: Amazon Reviews, MovieLens25M, and MovieLens1M datasets. In general, our results confirm that multimodal representations are able to boost the performance of unimodal models from 0.919 to 0.969 of accuracy on Amazon Reviews and 0.907 to 0.918 of AUC on MovieLens25M. However, experiments on both MovieLens datasets indicate the importance of the meaningful input data to the given task. In this article, we show that the choice of the technique for building multimodal representation is crucial to obtain the highest possible model's performance, that comes with the proper modalities combination. Such choice relies on: the influence that each modality has on the analyzed machine learning (ML) problem; the type of the ML task; the memory constraints while training and predicting phase.

Cornell University - arXiv, Sep 2, 2022

This paper presents research on a prototype developed to serve the quantitative study of public p... more This paper presents research on a prototype developed to serve the quantitative study of public policy design. This sub-discipline of political science focuses on identifying actors, relations between them, and tools at their disposal in health, environmental, economic, and other policies. Our system aims to automate the process of gathering legal documents, annotating them with Institutional Grammar, and using hypergraphs to analyse interrelations between crucial entities. Our system is tested against the UNESCO Convention for the Safeguarding of the Intangible Cultural Heritage from 2003, a legal document regulating essential aspects of international relations securing cultural heritage.

Food Computing is currently a fast-growing field of research. Natural language processing (NLP) i... more Food Computing is currently a fast-growing field of research. Natural language processing (NLP) is also increasingly essential in this field, especially for recognising food entities. However, there are still only a few well-defined tasks that serve as benchmarks for solutions in this area. We introduce a new dataset – called TASTEset – to bridge this gap. In this dataset, Named Entity Recognition (NER) models are expected to find or infer various types of entities helpful in processing recipes, e.g. food products, quantities and their units, names of cooking processes, physical quality of ingredients, their purpose, taste. The dataset consists of 700 recipes with more than 13,000 entities to extract. We provide a few state-of-the-art baselines of named entity recognition models, which show that our dataset poses a solid challenge to existing models. The best model achieved, on average, 0.95 F1 score, depending on the entity type – from 0.781 to 0.982. We share the dataset and the t...

Electronics

Recommendation systems have lately been popularised globally. However, often they need to be adap... more Recommendation systems have lately been popularised globally. However, often they need to be adapted to particular data and the use case. We have developed a machine learning-based recommendation system, which can be easily applied to almost any items and/or actions domain. Contrary to existing recommendation systems, our system supports multiple types of interaction data with various modalities of metadata through a multi-modal fusion of different data representations. We deployed the system into numerous e-commerce stores, e.g., food and beverages, shoes, fashion items, and telecom operators. We present our system and its main algorithms for data representations and multi-modal fusion. We show benchmark results on open datasets that outperform the state-of-the-art prior work. We also demonstrate use cases for different e-commerce sites.

Despite recent breakthroughs in Machine Learning for Natural Language Processing, the Natural Lan... more Despite recent breakthroughs in Machine Learning for Natural Language Processing, the Natural Language Inference (NLI) problems still constitute a challenge. To this purpose we contribute a new dataset that focuses exclusively on the factivity phenomenon; however, our task remains the same as other NLI tasks, i.e. prediction of entailment, contradiction or neutral (ECN). The dataset contains entirely natural language utterances in Polish and gathers 2,432 verb-complement pairs and 309 unique verbs. The dataset is based on the National Corpus of Polish (NKJP) and is a representative sample in regards to frequency of main verbs and other linguistic features (e.g. occurrence of internal negation). We found that transformer BERT-based models working on sentences obtained relatively good results ($\approx89\%$ F1 score). Even though better results were achieved using linguistic features ($\approx91\%$ F1 score), this model requires more human labour (humans in the loop) because features ...

Communications in Computer and Information Science, 2020

Fashion retrieval is the challenging task of finding an exact match for fashion items contained w... more Fashion retrieval is the challenging task of finding an exact match for fashion items contained within an image. Difficulties arise from the fine-grained nature of clothing items, very large intra-class and inter-class variance. Additionally, query and source images for the task usually come from different domains-street photos and catalogue photos respectively. Due to these differences, a significant gap in quality, lighting, contrast, background clutter and item presentation exists between domains. As a result, fashion retrieval is an active field of research both in academia and the industry. Inspired by recent advancements in Person Re-Identification research, we adapt leading ReID models to be used in fashion retrieval tasks. We introduce a simple baseline model for fashion retrieval, significantly outperforming previous state-of-the-art results despite a much simpler architecture. We conduct in-depth experiments on Street2Shop and DeepFashion datasets and validate our results. Finally, we propose a cross-domain (cross-dataset) evaluation method to test the robustness of fashion retrieval models.

2018 Baltic URSI Symposium (URSI), 2018

Offer images play an important role for e-commerce platforms in communicating features that are h... more Offer images play an important role for e-commerce platforms in communicating features that are hard to express with plain text. Images affect consumer's attitude and emotions toward particular products and the whole platform quality. This paper makes contribution to show the way of providing measurable indicators for product images business quality. The term business quality means ‘pure’ item visualisation giving easy to capture information about products of each offer. Additionally, this paper exposes the superiority of deep learning techniques over standard ones used to detect information in images.

Breast cancer is one of the most dangerous tumors for middle-aged and older women, and mammograph... more Breast cancer is one of the most dangerous tumors for middle-aged and older women, and mammography is its most reliable early detection method. In this paper, a fully automated method for detection of mass-like objects is proposed. The main stage of the algorithm is non-linear histogram conversion based on Rayleigh transformation. That approach gives us mammograms with significantly improved masses visibility and, thus, easier way to segment a potential mass object. Achieved results confirm the usefulness proposed method for application in mammography-oriented content-based image retrieval system and are comparable to other state-of art methods. Słowa kluczowe: mammografia, detekcja guzów, nieliniowa korekcja histogramu 1. Wprowadzenie Najczęściej występującym w Polsce nowotworem złośliwym u kobiet jest rak piersi, który stanowi około 20% wszystkich zachorowań na nowotwory [1]. Jest on równocześnie przyczyną największej liczby zgonów wywołanych przez nowotwory złośliwe u kobiet. W c...

Advances in Intelligent Systems and Computing, 2013

The paper desribes a novel approach to discovery of offers in multi-commodity trade on open marke... more The paper desribes a novel approach to discovery of offers in multi-commodity trade on open markets. Methods for gathering user requirements and for offer search are described. The process utilizes semantic technologies and multi-agent architecture. Examplex are shown in the domain of trading electric energy with the use of M 3 ontology.

The paper describes a lexical layer for ontologies and scenarios of its population. This layer ex... more The paper describes a lexical layer for ontologies and scenarios of its population. This layer extends lexical descriptions of the given ontology. It defines terms and their lexicalized meanings (given with contexts) associated with elements in the ontology. Additionally, it provides links to commonly used lexical knowledge resources.W artykule jest opisana warstwa leksykalna (słownikowa) ontologii oraz scenariusze jej budowy. Ta warstwa rozszerza opisy słowne danej ontologii, definiuje znaczenia leksykalne za pomocą słów występujących w kontekście danego terminu oraz wiąże je z elementami ontologii. Możliwe jest dołączenie znaczeń leksykalnych zdefiniowanych za pomocą znanych zasobów, jak WordNet czy Wikipedia

Sensors, 2022

This paper describes a notion of substitutions in food recipes and their ontology design pattern.... more This paper describes a notion of substitutions in food recipes and their ontology design pattern. We build upon state-of-the-art models for food and process. We also present scenarios and examples for the design pattern. Finally, the pattern is mapped to available and relevant domain ontologies and made publicly available at the ontologydesignpatterns.org portal.

Neural Information Processing, 2021

Biocybernetics and Biomedical Engineering, 2008

Although mammography is a standard of reference for detection of early breast cancer, as many as ... more Although mammography is a standard of reference for detection of early breast cancer, as many as 25% of breast cancers may be missed. To reduce the possibility of missing a cancer, the following methods and tools have been proposed: continuing education and training, prospective double reading, retrospective evaluation of missed cases, and use of computer-aided detection (CAD). The purpose of the reported work was to evaluate the usefulness and the potential of our aiding tools: an ontology driven editor for mammographic lesion description (MammoEdit) and a CAD-tool (MammoViewer) to enhance radiologist’s diagnostic performance. To this end test sample of mammograms was analyzed twice, without and with aiding tools. The obtained data were analyzed using (ROC) analysis and Kappa statistics. Statistical analysis of the test data demonstrated potential of both tools to enhance radiologist’s diagnostic performance.

ArXiv, 2020

State-of-the-art solutions for Natural Language Processing (NLP) are able to capture a broad rang... more State-of-the-art solutions for Natural Language Processing (NLP) are able to capture a broad range of contexts, like the sentence-level context or document-level context for short documents. But these solutions are still struggling when it comes to longer, real-world documents with the information encoded in the spatial structure of the document, such as page elements like tables, forms, headers, openings or footers; complex page layout or presence of multiple pages. To encourage progress on deeper and more complex Information Extraction (IE) we introduce a new task (named Kleister) with two new datasets. Utilizing both textual and structural layout features, an NLP system must find the most important information, about various types of entities, in long formal documents. We propose Pipeline method as a text-only baseline with different Named Entity Recognition architectures (Flair, BERT, RoBERTa). Moreover, we checked the most popular PDF processing tools for text extraction (pdf2d...

Document Analysis and Recognition – ICDAR 2021, 2021

The relevance of the Key Information Extraction (KIE) task is increasingly important in natural l... more The relevance of the Key Information Extraction (KIE) task is increasingly important in natural language processing problems. But there are still only a few well-defined problems that serve as benchmarks for solutions in this area. To bridge this gap, we introduce two new datasets (Kleister NDA and Kleister Charity). They involve a mix of scanned and born-digital long formal English-language documents. In these datasets, an NLP system is expected to find or infer various types of entities by employing both textual and structural layout features. The Kleister Charity dataset consists of 2,788 annual financial reports of charity organizations, with 61,643 unique pages and 21,612 entities to extract. The Kleister NDA dataset has 540 Non-disclosure Agreements, with 3,229 unique pages and 2,160 entities to extract. We provide several state-of-the-art baseline systems from the KIE domain (Flair, BERT, RoBERTa, LayoutLM, LAMBERT), which show that our datasets pose a strong challenge to existing models. The best model achieved an 81.77 % and an 83.57 % F1-score on respectively the Kleister NDA and the Kleister Charity datasets. We share the datasets to encourage progress on more in-depth and complex information extraction tasks.

Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL), 2019

Recent developments in Named Entity Recognition (NER) have resulted in better and better models. ... more Recent developments in Named Entity Recognition (NER) have resulted in better and better models. However, is there a glass ceiling? Do we know which types of errors are still hard or even impossible to correct? In this paper, we present a detailed analysis of the types of errors in state-of-the-art machine learning (ML) methods. Our study reveals the weak and strong points of the Stanford, CMU, FLAIR, ELMO and BERT models, as well as their shared limitations. We also introduce new techniques for improving annotation, for training processes and for checking a model's quality and stability. Presented results are based on the CoNLL 2003 data set for the English language. A new enriched semantic annotation of errors for this data set and new diagnostic data sets are attached in the supplementary materials. • (Manning, 2011) on linguistic limitations in building a perfect Part-of-Speech Tagger.

Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, 2018

Datasets that boosted state-of-the-art solutions for Question Answering (QA) systems prove that i... more Datasets that boosted state-of-the-art solutions for Question Answering (QA) systems prove that it is possible to ask questions in natural language manner. However, users are still used to query-like systems where they type in keywords to search for answer. In this study we validate which parts of questions are essential for obtaining valid answer. In order to conclude that, we take advantage of LIME-a framework that explains prediction by local approximation. We find that grammar and natural language is disregarded by QA. Stateof-the-art model can answer properly even if 'asked' only with a few words with high coefficients calculated with LIME. According to our knowledge, it is the first time that QA model is being explained by LIME.