Browse
Search Research Outputs
Recent Additions
- PublicationГүн сургалтын арга ашигласан Монгол дохионы хэлний хөрвүүлэгч(2022-11-04)Энэ ажлаар Монгол дохионы хэлний анхны орчуулагчийг хүний онцлог цэгүүдийг ашигласан гүн сур- галтын аргаар бий болгов. Компьютер дүрс боловсруулалтын салбарт энэ төрлийн асуудал дээр гүн сур- галтын арга хэрэглэж сургахад их хэмжээний өгөгдөл шаардлагатай байдаг. Одоогоор дэлхийд 300 гаруй дохионы хэл байдаг буюу улс бүр өөрсдийн дохионы хэлтэй байдаг. Бид энэ ажлын хүрээнд Монгол хэлний анхны дохионы хэлний өгөгдлийг үүсгэж ашигласан болно. Бид эхний ээлжинд 10 өгүүлбэрийг илэрхийлсэн дохионы хэлний өндөр чанартай 869 видеог бэлтгэн үүнээс хүний нүүр, цээж, баруун, зүүн га- раас нийтдээ 1662 ширхэг хүний биеийн онцлог цэгүүдийг ашиглан сургасан гүн сургалтын загвар гаргаж авав. Бидний сургасан загвар 96% нарийвчлалтайгаар дохионы хэлийг зөв таньж орчуулдаг болсон.
- PublicationCoarse-to-Fine Evolutionary Method for Fast Horizon Detection in Maritime Images(2021-12-01)Horizon detection is useful in maritime image processing for various purposes, such as estimation of camera orientation, registration of consecutive frames, and restriction of the object search region. Existing horizon detection methods are based on edge extraction. For accuracy, they use multiple images, which are filtered with different filter sizes. However, this increases the processing time. In addition, these methods are not robust to blurting. Therefore, we developed a horizon detection method without extracting the candidates from the edge information by formulating the horizon detection problem as a global optimization problem. A horizon line in an image plane was represented by two parameters, which were optimized by an evolutionary algorithm (genetic algorithm). Thus, the local and global features of a horizon were concurrently utilized in the optimization process, which was accelerated by applying a coarse-to-fine strategy. As a result, we could detect the horizon line on high-resolution maritime images in about 50ms. The performance of the proposed method was tested on 49 videos of the Singapore marine dataset and the Buoy dataset, which contain over 16000 frames under different scenarios. Experimental results show that the proposed method can achieve higher accuracy than state-of-the-art methods.
- PublicationAllocation of humanitarian aid after a weather disaster(2023)This paper tests whether need or political economy factors determine the allocation of humanitarian aid in the wake of the 2015/16 winter disaster in Mongolia. The identification strategy exploits the exogenous nature of the extremely cold, snowy winter and its spatial variation across Mongolia as well as the fact that the Government defined clear criteria of need across districts based on meteorological risk projections. Using rich district-level data, we distinguish between humanitarian aid delivered by the Mongolian Government and by international donors at the extensive margin (whether a district received any aid) and intensive margin (targeted households per district). Results show that projected need is the strongest predictor for the allocation of international humanitarian aid across districts. Projected need is less relevant for the allocation of governmental humanitarian aid. We do not find evidence that political alignment or core voter considerations matter for either governmental or international humanitarian aid in this young democracy. � 2023 Elsevier Ltd
- PublicationAn Architecture and a Methodology Enabling Interoperability within and across Universities(2022)We propose a general methodology and an infrastructure which allows to achieve interoperability within the same university and across universities. The former goal is achieved by incrementally defining and building a knowledge graph (KG) using data coming from multiple heterogeneous databases. Interoperability across universities is achieved by having a reference KG schema that each university can adapt to the local needs, but keeping track of the changes, and by natively supporting multilinguality. We achieve this latter requirement by exploiting a multilingual lexical resource containing more than one thousand languages and by seamlessly translating across the schemas and also (to some extent) across the data written in the local languages. The effectiveness of the proposed approach is proven by the services developed in the context of two different projects conducted in two universities in Italy and Mongolia.
- PublicationGeneralization and Personalization of Mobile Sensing-Based Mood Inference Models(2022-12)Mood inference with mobile sensing data has been studied in ubicomp literature over the last decade. This inference enables context-aware and personalized user experiences in general mobile apps and valuable feedback and interventions in mobile health apps. However, even though model generalization issues have been highlighted in many studies, the focus has always been on improving the accuracies of models using different sensing modalities and machine learning techniques, with datasets collected in homogeneous populations. In contrast, less attention has been given to studying the performance of mood inference models to assess whether models generalize to new countries. In this study, we collected a mobile sensing dataset with 329K self-reports from 678 participants in eight countries (China, Denmark, India, Italy, Mexico, Mongolia, Paraguay, UK) to assess the effect of geographical diversity on mood inference models. We define and evaluate country-specific (trained and tested within a country), continent-specific (trained and tested within a continent), country-agnostic (tested on a country not seen on training data), and multi-country (trained and tested with multiple countries) approaches trained on sensor data for two mood inference tasks with population-level (non-personalized) and hybrid (partially personalized) models. We show that partially personalized country-specific models perform the best yielding area under the receiver operating characteristic curve (AUROC) scores of the range 0.78--0.98 for two-class (negative vs. positive valence) and 0.76--0.94 for three-class (negative vs. neutral vs. positive valence) inference. Further, with the country-agnostic approach, we show that models do not perform well compared to country-specific settings, even when models are partially personalized. We also show that continent-specific models outperform multi-country models in the case of Europe. Overall, we uncover generalization issues of mood inference models to new countries and how the geographical similarity of countries might impact mood inference.
Most viewed
- PublicationGeneralization and Personalization of Mobile Sensing-Based Mood Inference Models(2022-12)Mood inference with mobile sensing data has been studied in ubicomp literature over the last decade. This inference enables context-aware and personalized user experiences in general mobile apps and valuable feedback and interventions in mobile health apps. However, even though model generalization issues have been highlighted in many studies, the focus has always been on improving the accuracies of models using different sensing modalities and machine learning techniques, with datasets collected in homogeneous populations. In contrast, less attention has been given to studying the performance of mood inference models to assess whether models generalize to new countries. In this study, we collected a mobile sensing dataset with 329K self-reports from 678 participants in eight countries (China, Denmark, India, Italy, Mexico, Mongolia, Paraguay, UK) to assess the effect of geographical diversity on mood inference models. We define and evaluate country-specific (trained and tested within a country), continent-specific (trained and tested within a continent), country-agnostic (tested on a country not seen on training data), and multi-country (trained and tested with multiple countries) approaches trained on sensor data for two mood inference tasks with population-level (non-personalized) and hybrid (partially personalized) models. We show that partially personalized country-specific models perform the best yielding area under the receiver operating characteristic curve (AUROC) scores of the range 0.78--0.98 for two-class (negative vs. positive valence) and 0.76--0.94 for three-class (negative vs. neutral vs. positive valence) inference. Further, with the country-agnostic approach, we show that models do not perform well compared to country-specific settings, even when models are partially personalized. We also show that continent-specific models outperform multi-country models in the case of Europe. Overall, we uncover generalization issues of mood inference models to new countries and how the geographical similarity of countries might impact mood inference.
- PublicationAllocation of humanitarian aid after a weather disaster(2023)This paper tests whether need or political economy factors determine the allocation of humanitarian aid in the wake of the 2015/16 winter disaster in Mongolia. The identification strategy exploits the exogenous nature of the extremely cold, snowy winter and its spatial variation across Mongolia as well as the fact that the Government defined clear criteria of need across districts based on meteorological risk projections. Using rich district-level data, we distinguish between humanitarian aid delivered by the Mongolian Government and by international donors at the extensive margin (whether a district received any aid) and intensive margin (targeted households per district). Results show that projected need is the strongest predictor for the allocation of international humanitarian aid across districts. Projected need is less relevant for the allocation of governmental humanitarian aid. We do not find evidence that political alignment or core voter considerations matter for either governmental or international humanitarian aid in this young democracy. � 2023 Elsevier Ltd
- PublicationPageRank алгоритмаар өгүүлбэр дэх ижил үгийн утгыг тогтоох нь(2019-05)Өгүүллээр монгол хэлний "ижил үг" буюу бичлэг ижил ч утга нь өөр үгс өгүүлбэрт аль утгаараа орсныг графт суурилсан статистик аргаар тодорхойлохыг оролдлоо. Бид 1.6 сая орчим үгтэй гар тэмдэглэгээт хөмрөгөөс 440 мянга орчим оройтой утгын граф байгуулсан ба уг графаас туршилтын хөмрөгт орсон ижил үгсийн утгыг хайхад 55.4 хувь зөв тодорхойлсон. Энэ нь олон улсын жишигтэй харьцуулахуйц үр дүн юм.
- PublicationAn Architecture and a Methodology Enabling Interoperability within and across Universities(2022)We propose a general methodology and an infrastructure which allows to achieve interoperability within the same university and across universities. The former goal is achieved by incrementally defining and building a knowledge graph (KG) using data coming from multiple heterogeneous databases. Interoperability across universities is achieved by having a reference KG schema that each university can adapt to the local needs, but keeping track of the changes, and by natively supporting multilinguality. We achieve this latter requirement by exploiting a multilingual lexical resource containing more than one thousand languages and by seamlessly translating across the schemas and also (to some extent) across the data written in the local languages. The effectiveness of the proposed approach is proven by the services developed in the context of two different projects conducted in two universities in Italy and Mongolia.
- PublicationГүн сургалтын арга ашигласан Монгол дохионы хэлний хөрвүүлэгч(2022-11-04)Энэ ажлаар Монгол дохионы хэлний анхны орчуулагчийг хүний онцлог цэгүүдийг ашигласан гүн сур- галтын аргаар бий болгов. Компьютер дүрс боловсруулалтын салбарт энэ төрлийн асуудал дээр гүн сур- галтын арга хэрэглэж сургахад их хэмжээний өгөгдөл шаардлагатай байдаг. Одоогоор дэлхийд 300 гаруй дохионы хэл байдаг буюу улс бүр өөрсдийн дохионы хэлтэй байдаг. Бид энэ ажлын хүрээнд Монгол хэлний анхны дохионы хэлний өгөгдлийг үүсгэж ашигласан болно. Бид эхний ээлжинд 10 өгүүлбэрийг илэрхийлсэн дохионы хэлний өндөр чанартай 869 видеог бэлтгэн үүнээс хүний нүүр, цээж, баруун, зүүн га- раас нийтдээ 1662 ширхэг хүний биеийн онцлог цэгүүдийг ашиглан сургасан гүн сургалтын загвар гаргаж авав. Бидний сургасан загвар 96% нарийвчлалтайгаар дохионы хэлийг зөв таньж орчуулдаг болсон.