An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

The Self-Organizing Map

Please provide a short abstract of 100 to 250 words. The abstract should not contain any undefined abbreviations or unspecified references. Work planned but not completed should not appear in the abstract.

User Modeling and User-Adapted Interaction

Recently, the character-word lattice structure has been proved to be effective for Chinese named entity recognition (NER) by incorporating the word information. However, since the lattice structure is complex and dynamic, the lattice-based models are hard to fully utilize the parallel computation of GPUs and usually have a low inference speed. In this paper, we propose FLAT: Flat-LAttice Transformer for Chinese NER, which converts the lattice structure into a flat structure consisting of spans. Each span corresponds to a character or latent word and its position in the original lattice. With the power of Transformer and well-designed position encoding, FLAT can fully leverage the lattice information and has an excellent parallel ability. Experiments on four datasets show FLAT outperforms other lexicon-based models in performance and efficiency.

/pdf/flat-chinese-ner-using-flat-lattice-transformer-3zqhim4v4f.pdf

FLAT: Chinese NER Using Flat-Lattice Transformer

http://cgi.di.uoa.gr/~stzikop/docs/Journal_Tzikopoulos_2011.pdf

A fully automated scheme for mammographic segmentation and classification based on breast density and asymmetry

In this paper, we propose a novel end-to-end approach for scalable visual search infrastructure. We discuss the challenges we faced for a massive volatile inventory like at eBay and present our solution to overcome those. We harness the availability of large image collection of eBay listings and state-of-the-art deep learning techniques to perform visual search at scale. Supervised approach for optimized search limited to top predicted categories and also for compact binary signature are key to scale up without compromising accuracy and precision. Both use a common deep neural network requiring only a single forward inference. The system architecture is presented with in-depth discussions of its basic components and optimizations for a trade-off between search relevance and latency. This solution is currently deployed in a distributed cloud infrastructure and fuels visual search in eBay ShopBot and Close5. We show benchmark on ImageNet dataset on which our approach is faster and more accurate than several unsupervised baselines. We share our learnings with the hope that visual search becomes a first class citizen for all large scale search engines rather than an afterthought.

Visual Search at eBay

Microcalcifications are one of more important signs enabling detection of breast cancer at early stage. The main goal of the research was designing and realization of a system for automatic detection and classification of microcalcifications, taking advantage of proposed automatic feature selection algorithm. The first step of the detection algorithm is to segment the individual objects: potential microcalcifications. This is achieved by applying opening by reconstruction top-hat technique and image thresholding based on approximation of an image local histogram with a probability density function of Gauss distribution. Selected features of the segmented objects are used as inputs to neural networks. The first classifier verifies the initial detection and the others assess a diagnosis of the input objects. The algorithm results are locations of suggested microcalcifications and optionally automatic diagnosis. The presented form of the system was verified in clinical tests using diagnosed databases (DDSM from University of South Florida and own digitized database of mammograms ). Achieved results are promising, comparable with other known systems. Efficiency of microcalcifications detection was up to 90%.

Segmentation and feature extraction for reliable classification of microcalcifications in digital mammograms

In this study the peels of ecologically grown apple (Malus domestica) cultivars: Gold Milenium (a new scab‐resistant variety) and Papierowka (Papirovka; an old, sensitive to apple scab variety) were examined for their composition (phenolic compounds, triterpenoids, simple organic acids, macro-, microelements, reducing sugars, l-ascorbic acid), pro- and antioxidant properties as well as their application in reduction of the oxidative stress in cultured human skin fibroblast. The higher content of phenolic compounds correlated with the greater pro- and antioxidant activity of the peels of Papierowka compared to Gold Milenium in DPPH·, ABTS+, FRAP and CUPRAC assays as well as an ability to inhibition of lipid peroxidation. The quantity of the compounds strongly depended on the type of extraction. The extract of Papierowka peels possessed much higher amount of phenolic compounds compared to Gold Milenium (Papierowka: 3.68 ± 0.20 mg/g peel ultrasound assisted extraction (u.a.e); 2.02 ± 0.13 mg/g peel conventional extraction (c.e.); Gold Milenium: 1.46 ± 0.19 mg/g peel u.a.e; 1.15 ± 0.04 mg/g peel c.e. according the HPLC measurement). The pro-oxidant activity of the extract from Papierowka peels can be correlated with the content of phenolic compounds and metal ions as well. The apple peel extract is promising agent reducing the oxidative stress in skin fibroblast.

/pdf/phenolic-content-chemical-composition-and-anti-pro-oxidant-57wy8fzddt.pdf

Phenolic content, chemical composition and anti-/pro-oxidant activity of Gold Milenium and Papierowka apple peel extracts

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 (pdf2djvu, Tesseract and Textract) in order to analyze behavior of IE system in presence of errors introduced by these tools.

Kleister: A novel task for Information Extraction involving Long Documents with Complex Layout

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.

Kleister: Key Information Extraction Datasets Involving Long Documents with Complex Layouts

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 illustrates 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, training process, and for checking model quality and stability.

/pdf/named-entity-recognition-is-there-a-glass-ceiling-1f32jessb4.pdf

Anna Wróblewska

Papers

Segmentation and feature extraction for reliable classification of microcalcifications in digital mammograms

Phenolic content, chemical composition and anti-/pro-oxidant activity of Gold Milenium and Papierowka apple peel extracts

Kleister: A novel task for Information Extraction involving Long Documents with Complex Layout

Kleister: Key Information Extraction Datasets Involving Long Documents with Complex Layouts

Named Entity Recognition - Is There a Glass Ceiling?