Semantic Granularities Research Articles

Indonesian-English Textual Similarity Detection Using Universal Sentence Encoder (USE) and Facebook AI Similarity Search (FAISS)

The tremendous development in Natural Language Processing (NLP) has enabled the detection of bilingual and multilingual textual similarity. One of the main challenges of the Textual Similarity Detection (TSD) system lies in learning effective text representation. The research focuses on identifying similar texts between Indonesian and English across a broad range of semantic similarity spectrums. The primary challenge is generating English and Indonesian dense vector representation, a.k.a. embeddings that share a single vector space. Through trial and error, the research proposes using the Universal Sentence Encoder (USE) model to construct bilingual embeddings and FAISS to index the bilingual dataset. The comparison between query vectors and index vectors is done using two approaches: the heuristic comparison with Euclidian distance and a clustering algorithm, Approximate Nearest Neighbors (ANN). The system is tested with four different semantic granularities, two text granularities, and evaluation metrics with a cutoff value of k={2,10}. Four semantic granularities used are highly similar or near duplicate, Semantic Entailment (SE), Topically Related (TR), and Out of Topic (OOT), while the text granularities take on the sentence and paragraph levels. The experimental results demonstrate that the proposed system successfully ranks similar texts in different languages within the top ten. It has been proven by the highest F1@2 score of 0.96 for the near duplicate category on the sentence level. Unlike the near-duplicate category, the highest F1 scores of 0.77 and 0.89 are shown by the SE and TR categories, respectively. The experiment results also show a high correlation between text and semantic granularity.

Enhancing interaction representation for joint entity and relation extraction

Jointly extracting entities and their relations from texts is an important task in information extraction. Despite the great success, traditional models suffer from two problems. First, the same token embeddings are shared in two subtasks. It ignores the difference between semantic granularities, in which named entities are more dependent on local features and relations are semantic expressions relevant a whole sentence. Second, the interaction between two subtasks is rarely considered, which is important to encode semantic dependencies relevant to two named entities. To address the above problems, we presents a novel joint entity and relation extraction model. It constructs two independent token embedding modules for encoding features about entities and relations respectively. It enables to encode semantic representations with different granularities for named entities and entity relations. Then, a cross-attention is used to capture the interaction between two subtasks for learning semantic dependencies in a relation instance. The experimental results demonstrate that our model outperforms previous state-of-art models on several public datasets. Extensive additional experiments further confirm the effectiveness of our model.

Full-span named entity recognition with boundary regression

Span classification is a popular method for nested named entity recognition. To recognise full-span named entities, span-based models should enumerate and verify all possible entity spans in a sentence, which leads to serious problems regarding computational complexity and data imbalance. In this study, we propose a boundary regression model to support full-span named entity recognition, where a regression operation is adopted to refine spatial locations of entity spans in a sentence. Therefore, instead of exhaustively enumerating all possible spans, we need only verify a small number of them. Span boundaries are regressed to find all possible named entities in a sentence. Furthermore, for a better representation of long-named entities, a multi-granule sentence representation is adopted to encode semantic features with different semantic granularities. In our experiments, even enumerating a small number of entity spans, our model still has competitive performance, achieving 87.35% and 80.85% F1 scores on the ACE2005 and GENIA datasets. Analytical experiments show that our model is able to find all named entities in a sentence without exhaustively verifying all possible entity spans. It is effective in mitigating the computational complexity and data imbalance problems in full-span named entity recognition.

Towards POI-based large-scale land use modeling: spatial scale, semantic granularity, and geographic context

ABSTRACT The combination of spatial distribution, semantic characteristics, and sometimes temporal dynamics of POIs inside a geographic region can capture its unique land use characteristics. Most previous studies on POI-based land use modeling research focused on one geographic region and select one spatial scale and semantic granularity for land use characterization. There is a lack of understanding on the impact of spatial scale, semantic granularity, and geographic context on POI-based land use modeling, particularly large-scale land use modeling. In this study, we developed a scalable POI-based land use modeling framework and examined the impact of these three factors on POI-based land use characterization using data from three geographic regions. We developed a unified semantic representation framework for POI semantics that can help fuse heterogeneous POI data sources. Then, by combining POIs with a neural network language model, we developed a spatially explicit approach to learn the embedding representation of POIs and AOIs. We trained multiple supervised classifiers using AOI embeddings as input features to predict AOI land use at different semantic granularities. The classification performance of different land use classes was analyzed and compared across three geographic regions to identify the semantic representativeness of POI-based AOI embedding and the impact of geographic context.

Pre-training language model incorporating domain-specific heterogeneous knowledge into a unified representation

Existing technologies expand BERT from different perspectives, e.g. designing different pre-training tasks, different semantic granularities, and different model architectures. Few models consider expanding BERT from different text formats. In this paper, we propose a heterogeneous knowledge language model (HKLM), a unified pre-trained language model (PLM) for all forms of text, including unstructured text, semi-structured text, and well-structured text. To capture the corresponding relations among these multi-format knowledge, our approach uses masked language model objective to learn word knowledge, uses triple classification objective and title matching objective to learn entity knowledge and topic knowledge respectively. To obtain the aforementioned multi-format text, we construct a corpus in the tourism domain and conduct experiments on 5 tourism NLP datasets. The results show that our approach outperforms the pre-training of plain text using only 1/4 of the data. We further pre-train the domain-agnostic HKLM and achieve performance gains on the XNLI dataset.

High-quality domain expert finding method in CQA based on multi-granularity semantic analysis and interest drift

Expert finding is an important research field in community question answering (CQA). Traditional expert finding methods mainly exploit topic analysis and authority calculation methods to identify high-quality experts in certain fields. To avoid recommending questions to those experts who do not display the willingness or ability to provide high-quality answers, user interest drift and user quality should be considered. This study proposes a novel method named high-quality domain expert finding in CQA based on multi-granularity semantic analysis and interest drift (HQExpert). Firstly, HQExpert considers different semantic granularities by employing two models, a coarse-grained topic model LC-LDA and a fine-grained model (BERT), to capture the domain information of questions and users more accurately. Secondly, to address the diverse interests of the users, a user interest drift model in HQExpert is developed to dynamically represent the changes in the interests of the users at different periods. In addition, a user quality model is developed to further optimize the professional level of the user, finding experts who can provide high-quality answers and are interested in the current question. Finally, extensive experiments on two datasets from different domains demonstrate that the proposed HQExpert model can significantly improve the accuracy of finding high-quality experts.

Quality Assessment of Pre-Classification Maps Generated from Spaceborne/Airborne Multi-Spectral Images by the Satellite Image Automatic Mapper™ and Atmospheric/Topographic Correction™-Spectral Classification Software Products: Part 2 — Experimental Results

This paper complies with the Quality Assurance Framework for Earth Observation (QA4EO) international guidelines to provide a metrological/statistically-based quality assessment of the Spectral Classification of surface reflectance signatures (SPECL) secondary product, implemented within the popular Atmospheric/Topographic Correction (ATCOR™) commercial software suite, and of the Satellite Image Automatic Mapper™ (SIAM™) software product, proposed to the remote sensing (RS) community in recent years. The ATCOR™-SPECL and SIAM™ physical model-based expert systems are considered of potential interest to a wide RS audience: in operating mode, they require neither user-defined parameters nor training data samples to map, in near real-time, a spaceborne/airborne multi-spectral (MS) image into a discrete and finite set of (pre-attentional first-stage) spectral-based semi-concepts (e.g., “vegetation”), whose informative content is always equal or inferior to that of target (attentional second-stage) land cover (LC) concepts (e.g., “deciduous forest”). For the sake of simplicity, this paper is split into two: Part 1—Theory and Part 2—Experimental results. The Part 1 provides the present Part 2 with an interdisciplinary terminology and a theoretical background. To comply with the principle of statistics and the QA4EO guidelines discussed in the Part 1, the present Part 2 applies an original adaptation of a novel probability sampling protocol for thematic map quality assessment to the ATCOR™-SPECL and SIAM™ pre-classification maps, generated from three spaceborne/airborne MS test images. Collected metrological/ statistically-based quality indicators (QIs) comprise: (i) an original Categorical Variable Pair Similarity Index (CVPSI), capable of estimating the degree of match between a test pre-classification map’s legend and a reference LC map’s legend that do not coincide and must be harmonized (reconciled); (ii) pixel-based Thematic (symbolic, semantic) QIs (TQIs) and (iii) polygon-based sub-symbolic (non-semantic) Spatial QIs (SQIs), where all TQIs and SQIs are provided with a degree of uncertainty in measurement. Main experimental conclusions of the present Part 2 are the following. (I) Across the three test images, the CVPSI values of the SIAM™ pre-classification maps at the intermediate and fine semantic granularities are superior to those of the ATCOR™-SPECL single-granule maps. (II) TQIs of both the ATCOR™-SPECL and the SIAM™ tend to exceed community-agreed reference standards of accuracy. (III) Across the three test images and the SIAM™’s three semantic granularities, TQIs of the SIAM™ tend to be significantly higher (in statistical terms) than the ATCOR™-SPECL’s. Stemming from the proposed experimental evidence in support to theoretical considerations, the final conclusion of this paper is that, in compliance with the QA4EO objectives, the SIAM™ software product can be considered eligible for injecting prior spectral knowledge into the pre-attentive vision first stage of a novel generation of hybrid (combined deductive and inductive) RS image understanding systems, capable of transforming large-scale multi-source multi-resolution EO image databases into operational, comprehensive and timely knowledge/information products.

Multimedia event detection with multimodal feature fusion and temporal concept localization

We present a system for multimedia event detection. The developed system characterizes complex multimedia events based on a large array of multimodal features, and classifies unseen videos by effectively fusing diverse responses. We present three major technical innovations. First, we explore novel visual and audio features across multiple semantic granularities, including building, often in an unsupervised manner, mid-level and high-level features upon low-level features to enable semantic understanding. Second, we show a novel Latent SVM model which learns and localizes discriminative high-level concepts in cluttered video sequences. In addition to improving detection accuracy beyond existing approaches, it enables a unique summary for every retrieval by its use of high-level concepts and temporal evidence localization. The resulting summary provides some transparency into why the system classified the video as it did. Finally, we present novel fusion learning algorithms and our methodology to improve fusion learning under limited training data condition. Thorough evaluation on a large TRECVID MED 2011 dataset showcases the benefits of the presented system.

Context dependent semantic granularity

A fundamental issue to improve the accessibility to information resources is how to efficiently deal with huge amount(s) of data. In this respect, ontology driven techniques are expected to improve the overlap between the Cognitive Space applied by the user and the Information Space, which is defined by the information providers. In this paper we describe a powerful method to extract semantic granularities, which enable the navigation of a repository according to different levels of abstraction. In the formalisation we present, granularities are explicitly parameterised according to criteria induced by the context, which improves the method flexibility. Furthermore, the parameterisation assists the user allowing to formulate and refine the browsing criteria. Case studies are described to demonstrate how granularities ease the information sources browsing and to illustrate how they may vary according to the context. A validation of the cognitive principles behind the method is presented, together with the analysis of the results obtained by the experimentation.

Exploitation of semantic relationships and hierarchical data structures to support a user in his annotation and browsing activities in folksonomies

In this paper we present a new approach to supporting users to annotate and browse resources referred by a folksonomy. Our approach is characterized by the following novelties: (i) it proposes a probabilistic technique to quickly and accurately determine the similarity and the generalization degrees of two tags; (ii) it proposes two hierarchical structures and two related algorithms to arrange groups of semantically related tags in a hierarchy; this allows users to visualize tags of their interests according to desired semantic granularities and, then, helps them to find those tags best expressing their information needs. In this paper we first illustrate the technical characteristics of our approach; then we describe various experiments allowing its performance to be tested; finally, we compare it with other related approaches already proposed in the literature.

An HMM-based framework for video semantic analysis

Video semantic analysis is essential in video indexing and structuring. However, due to the lack of robust and generic algorithms, most of the existing works on semantic analysis are limited to specific domains. In this paper, we present a novel hidden Markove model (HMM)-based framework as a general solution to video semantic analysis. In the proposed framework, semantics in different granularities are mapped to a hierarchical model space, which is composed of detectors and connectors. In this manner, our model decomposes a complex analysis problem into simpler subproblems during the training process and automatically integrates those subproblems for recognition. The proposed framework is not only suitable for a broad range of applications, but also capable of modeling semantics in different semantic granularities. Additionally, we also present a new motion representation scheme, which is robust to different motion vector sources. The applications of the proposed framework in basketball event detection, soccer shot classification, and volleyball sequence analysis have demonstrated the effectiveness of the proposed framework on video semantic analysis.