| 12111. |
- Lindén, Johannes, 1993-, et al.
(author)
-
Bilingual Auto-Categorization Comparison of two LSTM Text Classifiers
- 2019
-
In: 2019 8th International Congress on Advanced Applied Informatics (IIAI-AAI). - 9781728126272
-
Conference paper (peer-reviewed)abstract
- Multi linguistic problems such as auto-categorization is not an easy task. It is possible to train different models for each language, another way to do auto-categorization is to build the model in one base language and use automatic translation from other languages to that base language. Different languages have a bias to a language specific grammar and syntax and will therefore pose problems to be expressed in other languages. Translating from one language into a non-verbal language could potentially have a positive impact of the categorization results. A non-verbal language could for example be pure information in form of a knowledge graph relation extraction from the text. In this article a comparison is conducted between Chinese and Swedish languages. Two categorization models are developed and validated on each dataset. The purpose is to make an auto-categorization model that works for n'importe quel langage. One model is built upon LSTM and optimized for Swedish and the other is an improved Bidirectional-LSTM Convolution model optimized for Chinese. The improved algorithm is trained on both languages and compared with the LSTM algorithm. The Bidirectional-LSTM algorithm performs approximately 20% units better than the LSTM algorithm, which is significant.
|
|
| 12112. |
- Lindén, Johannes, 1993-, et al.
(author)
-
Evaluating Combinations of Classification Algorithms and Paragraph Vectors for News Article Classification
- 2018
-
In: Proceedings of the 2018 Federated Conference on Computer Science and Information Systems. - Warzaw : Polskie Towarzystwo Informatyczne. - 9788394941970 ; , s. 489-495
-
Conference paper (peer-reviewed)abstract
- News companies have a need to automate and make the process of writing about popular and new events more effective. Current technologies involve robotic programs that fill in values in templates and website listeners that notify editors when changes are made so that the editor can read up on the source change on the actual website. Editors can provide news faster and better if directly provided with abstracts of the external sources and categorical meta-data that supports what the text is about. In this article, the focus is on the importance of evaluating critical parameter modifications of the four classification algorithms Decisiontree, Randomforest, Multi Layer perceptron and Long-Short-Term-Memory in a combination with the paragraph vector algorithms Distributed Memory and Distributed Bag of Words, with an aim to categorise news articles. The result shows that Decisiontree and Multi Layer perceptron are stable within a short interval, while Randomforest is more dependent on the parameters best split and number of trees. The most accurate model is Long-Short-Term-Memory model that achieves an accuracy of 71%.
|
|
| 12113. |
- Lindén, Johannes, 1993-
(author)
-
Extracting Text into Meta-Data : Improving machine text-understanding of news-media articles
- 2021
-
Licentiate thesis (other academic/artistic)abstract
- Society is constantly in need of information. It is important to consume event-based information of what is happening around us as well as facts and knowledge. As society grows, the amount of information to consume grows with it. This thesis demonstrates one way to extract and represent knowledge from text in a machine-readable way for news media articles. Three objectives are considered when developing a machine learning system to retrieve categories, entities, relations and other meta-data from text paragraphs. The first is to sort the terminology by topic; this makes it easier for machine learning algorithms to understand the text and the unique words used. The second objective is to construct a service for use in production, where scalability and performance are evaluated. Features are implemented to iteratively improve the model predictions, and several versions are run at the same time to, for example, compare them in an A/B test. The third objective is to further extract the gist of what is expressed in the text. The gist is extracted in the form of triples by connecting two related entities using a combination of natural language processing algorithms. The research presents a comparison between five different auto categorization algorithms, and an evaluation of their hyperparameters and how they would perform under the pressure of thousands of big, concurrent predictions. The aim is to build an auto-categorization system that can be used in the news media industry to help writers and journalists focus more on the story rather than filling in meta-data for each article. The best-performing algorithm is a Bidirectional Long-Short-Term-Memory neural network. Three different information extraction algorithms for extracting the gist of paragraphs are also compared. The proposed information extraction algorithm supports extracting information from texts in multiple languages with competitive accuracy compared with the state-of-the-art OpenIE and MinIE algorithms that can extract information in a single language. The use of the multi-linguistic models helps local-news media to write articles in different languages as a help to integrate immigrants into the society.
|
|
| 12114. |
- Lindén, Johannes, 1993-, et al.
(author)
-
Multi-language Information Extraction with Text Pattern Recognition
- 2021
-
In: Computer Science & Information Technology (CS & IT). - 9781925953541 ; , s. 1-17
-
Conference paper (peer-reviewed)abstract
- Information extraction is a task that can extract meta-data information from text. The research in this article proposes a new information extraction algorithm called GenerateIE. The proposed algorithm identifies pairs of entities and relations described in a piece of text. The extracted meta-data is useful in many areas, but within this research the focus is to use them in news-media contexts to provide the gist of the written articles for analytics and paraphrasing of news information. GenerateIE algorithm is compared with existing state of the art algorithms with two benefits. Firstly, the GenerateIE provides the co-referenced word as the entity instead of using he, she, it, etc. which is more beneficial for knowledge graphs. Secondly GenerateIE can be applied on multiple languages without changing the algorithm itself apart from the underlying natural language text-parsing. Furthermore, the performance of GenerateIE compared with state-of-the-art algorithms is not significantly better, but it offers competitive results.
|
|
| 12115. |
- Lindén, Johannes, 1993-, et al.
(author)
-
Productify news article classification model with Sagemaker
- 2020
-
In: Advances in Science, Technology and Engineering Systems. - : ASTES Journal. - 2415-6698. ; 5:2, s. 13-18
-
Journal article (peer-reviewed)abstract
- News companies have a need to automate and make the process of writing about popular and new events more effective. Current technologies involve robotic programs that fill in values in templates and website listeners that notify editors when changes are made so that the editor can read up on the source change on the actual website. Editors can provide news faster and better if directly provided with abstracts of the external sources and categorical meta-data that supports what the text is about. To make categorical meta-data a reality an auto-categorization model was created and optimized for Swedish articles written by local news journalists. The problem was that it was not scale-able enough to use out of the box. Instead of having this local model that could make good predictions of the text documents, the model is to be deployed in the cloud and an API interface is created. The API can be accessed from the tools where the articles is being written and therefore these services can automatically assign categories to the articles once the journalist is done writing it. To allow scale-ability to several thousands of simultaneously categorized articles and at the same time improving the workflow of deploying new models easier the API is uploaded to Sagemaker where several models are trained and once an improved model is found that model will be used in production in such a way that the system organically adapts to new written articles. An evaluation of Sagemaker API was done and it was concluded that the complexity of this solution was polynomial.
|
|
| 12116. |
- Linder, Juergen, et al.
(author)
-
Long-term sick-leavers with difficulty in resuming work : comparisons between psychiatric-somatic co-morbidity and monodiagnosis
- 2009
-
In: International Journal of Rehabilitation Research. - 0342-5282 .- 1473-5660. ; 32:1, s. 20-35
-
Journal article (peer-reviewed)abstract
- The number of patients with difficulty in resuming work after long-term sick leave has increased in several European countries including Sweden. The general aim of this study was a comprehensive description - based on multidisciplinary diagnostics and assessments - of patients with the common feature of marked difficulty in resuming working life after a long absence. A particular aim was to elucidate the possible effect of comorbidity on pain descriptors, disability, quality of life, assessed working ability and rehabilitation needs. Six hundred and thirty-five long-term sick leavers were referred from National Insurance Offices and consecutively accepted for investigation. Several self-report questionnaires were used. All patients were examined by three board-certified specialist physicians in psychiatry, orthopaedic surgery and rehabilitation medicine, respectively. Fifty-five percent of the patients had psychiatric-somatic comorbidity. The three most frequent combinations of diagnoses in the comorbidity group were fibromyalgia/myalgia and depressive episode, fibromyalgia/myalgia and recurrent depression, spinal pain and depressive episode, whereas the three most frequent in those with psychiatric diagnosis only were depressive episode, recurrent depression, phobias/anxiety. Differences in pain descriptors and in difficulties with activities were found among the three groups. All had lower health-related quality of life than references. Only one-sixth had no assessed working capacity and only 3% were assessed as able to resume work without rehabilitation; 80% were multidisciplinarily assessed as needing rehabilitation. Patients with psychiatric diagnoses, with or without concomitant somatic diagnoses, need medical rehabilitation or medical/vocational rehabilitation in combination to a greater extent than patients with somatic diagnoses only. This implies that medical rehabilitation programmes ought to adapt increasingly to the needs of patients with psychiatric-somatic comorbidity.
|
|
| 12117. |
- Linder, J, et al.
(author)
-
Long-term sick-leavers with fibromyalgia : Comparing their multidisciplinarily assessed characteristics with those of others with chronic pain conditions and depression
- 2009
-
In: Journal of Multidisciplinary Healthcare. - 1178-2390 .- 1178-2390. ; 2, s. 23-37
-
Journal article (peer-reviewed)abstract
- Objective: The aim was to gain knowledge of fibromyalgia (FM) patients on long-term sick leave and with particular difficulties in resuming work, and to compare them with patients with myalgia, back or joint diagnoses, and depression.Methods: Patients were identified by and referred from social insurance offices and were multidisciplinarily examined by three board-certified specialists in psychiatry, orthopedic surgery and rehabilitation medicine. Ninety-two women were diagnosed with FM only. Three female comparison groups were chosen: depression, back/joint diagnoses, and myalgia.Results and conclusions: Ceaseless pain was reported by 73% of FM patients, 54% of back/joint diagnoses patients, 43% of myalgia patients, and 35% of depression patients. The distribution of pain (>50%) in FM patients was to almost all regions of the body, and in depression patients to the lower dorsal neck, upper shoulders and lumbosacral back but not in the anterior body. Reduced sleep was more evident in FM patients. FM patients did not meet more criteria for personality disorder than patients with the other somatic pain conditions. The most common dimension of “personality traits” of somatic pain conditions was the “obsessive compulsive” but at a level clearly below that indicating a personality disorder. More FM patients experienced disabilities, the most common being in the mobility and domestic-life areas.
|
|
| 12118. |
- Linder, Jurgen, et al.
(author)
-
RELATIONSHIP BETWEEN SLEEP DISTURBANCE, PAIN, DEPRESSION AND FUNCTIONING IN LONG-TERM SICK-LISTED PATIENTS EXPERIENCING DIFFICULTY IN RESUMING WORK
- 2014
-
In: JOURNAL OF REHABILITATION MEDICINE. - : Medical Journals Sweden AB. - 1651-2081 .- 1650-1977. ; 46:8, s. 798-805
-
Journal article (peer-reviewed)abstract
- Objective: To describe the frequency of reported sleeping, depression and pain problems, the severity of these problems and the degree of self-estimated difficulties in mental functions and activities in relation to the sleep disturbance and pain category group in patients on long-term sick-leave. Design: Cross-sectional study. Patients: A total of 1206 patients experiencing difficulty in resuming work. Methods: Patient examinations by specialists in psychiatry, orthopaedic surgery and rehabilitation medicine. Validated questionnaires, including status regarding depression, sleep, pain and functioning were used. Results: The prevalence of sleep disturbance was 83%; 74% of the patients with moderate/severe sleep disturbance also had moderate/severe pain problems and 26% had no/mild pain problems. Fifty-seven percent of the patients with no/mild sleep disturbance and 83% of the patients with moderate/severe sleep disturbance also had depression. The degree of difficulty in performing the 6 selected International Classification of Functioning, Disability and Health activities and mental functions was higher for the category with moderate/severe sleep problems, compared with those with no/mild sleep problems. Conclusion: To optimize rehabilitation for patients on long-term sick-leave experiencing difficulties in returning to work, the results indicate a need also to focus attention on sleep problems and not only on pain and depression. This may entail the planning of measures to improve decision-making and concentration and alleviate lassitude, fatigability, sadness and pessimistic thoughts.
|
|
| 12119. |
- Linder, Tomas, et al.
(author)
-
Lateral light scattering in fibrous media
- 2013
-
In: Optics Express. - 1094-4087. ; 21:6, s. 7835-7840
-
Journal article (peer-reviewed)abstract
- Lateral light scattering in fibrous media is investigated by computing the modulation transfer function (MTF) of 22 paper samples using a Monte Carlo model. The simulation tool uses phase functions from infinitely long homogenous cylinders and the directional inhomogeneity of paper is achieved by aligning the cylinders in the plane. The inverse frequency at half maximum of the MTF is compared to both measurements and previous simulations with isotropic and strongly forward single scattering phase functions. It is found that the conical scattering by cylinders enhances the lateral scattering and therefore predicts a larger extent of lateral light scattering than models using rotationally invariant single scattering phase functions. However, it does not fully reach the levels of lateral scattering observed in measurements. It is argued that the hollow lumen of a wood fiber or dependent scattering effects must be considered for a complete description of lateral light scattering in paper.
|
|
| 12120. |
- Linderborg, Otto, 1985-
(author)
-
Classical Greek Political Science in Action : Social Critique in Thucydides’ Mytilenean Debate and Melian Dialogue
- 2021
-
In: Journal of Greco-Roman Studies. - : The Korean Society of Greco-Roman Studies. - 1225-1828. ; 60:1, s. 17-35
-
Journal article (peer-reviewed)abstract
- This article brings together two central episodes in Thucydides’ History of the Peloponnesian War. The interpreted episodes are the Mytilenean Debate in Book III and the Melian Dialogue in Book V of the History. In the present study, these episodes are approached as original inquiries into moral and political matters, assuming the shape of subversive social criticism: immanent critique. A particular focus lies on the socio-political ordering principles scrutinized in the Thucydidean episodes. In the Mytilenean Debate, it is the principle of expediency (τὸ ξύμφορον) that is given the upper hand, whereas in the Melian dialogue the dominating social ordering principle is that of safety and survival (σωτηρία). In each episode, a contending point of view aims at undermining the pre-eminence of the stronger principle. However, the critique only succeeds if the subversion is managed from within, and if it pays outward allegiance to the frames determined by the supreme communal will.
|
|
