| 1. |
- Boman, Magnus, et al.
(författare)
-
Learning Machines
- 2018
-
Ingår i: <em>Learning, Inference and Control of Multi-Agent Systems</em>. ; , s. 610-613
-
Konferensbidrag (refereegranskat)
|
|
| 2. |
- Boman, Magnus, et al.
(författare)
-
Learning machines in Internet-delivered psychological treatment
- 2019
-
Ingår i: Progress in Artificial Intelligence. - : Springer Verlag. - 2192-6352 .- 2192-6360. ; 8:4, s. 475-485
-
Tidskriftsartikel (refereegranskat)abstract
- A learning machine, in the form of a gating network that governs a finite number of different machine learning methods, is described at the conceptual level with examples of concrete prediction subtasks. A historical data set with data from over 5000 patients in Internet-based psychological treatment will be used to equip healthcare staff with decision support for questions pertaining to ongoing and future cases in clinical care for depression, social anxiety, and panic disorder. The organizational knowledge graph is used to inform the weight adjustment of the gating network and for routing subtasks to the different methods employed locally for prediction. The result is an operational model for assisting therapists in their clinical work, about to be subjected to validation in a clinical trial.
|
|
| 3. |
|
|
| 4. |
- Ghoorchian, Kambiz, 1981-
(författare)
-
Graph Algorithms for Large-Scale and Dynamic Natural Language Processing
- 2019
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- In Natural Language Processing, researchers design and develop algorithms to enable machines to understand and analyze human language. These algorithms benefit multiple downstream applications including sentiment analysis, automatic translation, automatic question answering, and text summarization. Topic modeling is one such algorithm that solves the problem of categorizing documents into multiple groups with the goal of maximizing the intra-group document similarity. However, the manifestation of short texts like tweets, snippets, comments, and forum posts as the dominant source of text in our daily interactions and communications, as well as being the main medium for news reporting and dissemination, increases the complexity of the problem due to scalability, sparsity, and dynamicity. Scalability refers to the volume of the messages being generated, sparsity is related to the length of the messages, and dynamicity is associated with the ratio of changes in the content and topical structure of the messages (e.g., the emergence of new phrases). We improve the scalability and accuracy of Natural Language Processing algorithms from three perspectives, by leveraging on innovative graph modeling and graph partitioning algorithms, incremental dimensionality reduction techniques, and rich language modeling methods. We begin by presenting a solution for multiple disambiguation on short messages, as opposed to traditional single disambiguation. The solution proposes a simple graph representation model to present topical structures in the form of dense partitions in that graph and applies disambiguation by extracting those topical structures using an innovative distributed graph partitioning algorithm. Next, we develop a scalable topic modeling algorithm using a novel dense graph representation and an efficient graph partitioning algorithm. Then, we analyze the effect of temporal dimension to understand the dynamicity in online social networks and present a solution for geo-localization of users in Twitter using a hierarchical model that combines partitioning of the underlying social network graph with temporal categorization of the tweets. The results show the effect of temporal dynamicity on users’ spatial behavior. This result leads to design and development of a dynamic topic modeling solution, involving an online graph partitioning algorithm and a significantly stronger language modeling approach based on the skip-gram technique. The algorithm shows strong improvement on scalability and accuracy compared to the state-of-the-art models. Finally, we describe a dynamic graph-based representation learning algorithm that modifies the partitioning algorithm to develop a generalization of our previous work. A strong representation learning algorithm is proposed that can be used for extracting high quality distributed and continuous representations out of any sequential data with local and hierarchical structural properties similar to natural language text.
|
|
| 5. |
- Gogoulou, Evangelia, et al.
(författare)
-
Predicting treatment outcome from patient texts : The case of internet-based cognitive behavioural therapy
- 2021
-
Ingår i: EACL 2021 - 16th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference. - : Association for Computational Linguistics (ACL). - 9781954085022 ; , s. 575-580
-
Konferensbidrag (refereegranskat)abstract
- We investigate the feasibility of applying standard text categorisation methods to patient text in order to predict treatment outcome in Internet-based cognitive behavioural therapy. The data set is unique in its detail and size for regular care for depression, social anxiety, and panic disorder. Our results indicate that there is a signal in the depression data, albeit a weak one. We also perform terminological and sentiment analysis, which confirm those results. © 2021 Association for Computational Linguistics
|
|
| 6. |
- Hjelmblom, Magnus, 1973-
(författare)
-
Norm-Regulation of Agent Systems : Instrumentalizing an algebraic approach to agent system norms
- 2015
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- An architecture for norm-regulated multi-agent systems based on an algebraic approach to normative systems is instrumentalized and further developed. The core of the instrumentalization is a Prolog module, which together with a Java library can be used for creating client/server-based runtime systems. Norms are represented as conditional sentences, whose normative consequences are formulated by applying normative operators to descriptive conditions. From such general normative conditions follow normative sentences regarding specific states of affairs. These in turn result in permission or prohibition of individual actions in specific situations. Furthermore, an approach to turning runtime systems into instruments for problem-solving by using evolutionary mechanisms for evolving normative systems, is presented. The construction of norm-creating operators on conditions, which forms the basis for the representation of normative systems, is approached from two angles. (i) A logical analysis based on the Kanger-Lindahl theory of normative positions is conducted. This results in two extended sets of types of normative positions, and based on an algebraic version of one of these extended systems, a set of operators for creating agent-specific norms is constructed. (ii) An alternative analysis, which takes as its starting point a systematic exploration of types of state transitions, yields a set of norm-creating operators based on prohibition of transition types. It is furthermore argued that in the context of a class of transition systems, in which transitions are deterministic and associated with a single agent performing an act, operators based on (ii) specify a meaningful semantics of operators based on (i). Theoretical results together with shared code and example applications contribute to make possible theoretically sound, transparently described, and efficiently implemented norm-regulated autonomous agent systems.
|
|
| 7. |
- Xing, K. Z., et al.
(författare)
-
The electronic and geometric structures of neutral and potassium-doped poly[3-(4-octylphenyl)thiophene] studied by photoelectron spectroscopy
- 1996
-
Ingår i: Synthetic metals. - : Elsevier. - 0379-6779 .- 1879-3290. ; 76:1-3, s. 263-267
-
Tidskriftsartikel (refereegranskat)abstract
- The electronic and geometric structures of poly [3-(4-octylphenyl)thiophene] have been studied by X-ray and ultraviolet photoelectron spectroscopy (XPS and UPS, respectively). Thermochromic effects, and new charge induced states generated by potassium doping, have been observed by direct UPS measurements. The experimental results are in very good agreement with the results of theoretical quantum chemical calculations performed with the Austin Model 1 semi-empirical model and the valence-effective Hamiltonian pseudo-potential model.
|
|
| 8. |
- Xing, K. Z., et al.
(författare)
-
The electronic structure of neutral and alkali metal-doped poly[3-(4-octylphenyl)thiophene] studied by photoelectron spectroscopy
- 1996
-
Ingår i: Synthetic metals. - : Elsevier. - 0379-6779 .- 1879-3290. ; 80:1, s. 59-66
-
Tidskriftsartikel (refereegranskat)abstract
- The electronic structure of poly [3-(4-octylphenyl)thiophene] (POPT) has been studied by ultraviolet photoelectron spectroscopy (UPS) and X-ray photoelectron spectroscopy (XPS), as well as by quantum chemical calculations. Both temperature-dependent effects on the electronic structure of the neutral system, as well as the generation of new electronic states induced by doping with alkaline metals, have been observed. The experimental results are in good agreement with the results of the quantum chemical calculations.
|
|
| 9. |
- Aidas, Kestutis, et al.
(författare)
-
The Dalton quantum chemistry program system
- 2014
-
Ingår i: Wiley Interdisciplinary Reviews. Computational Molecular Science. - : Wiley. - 1759-0876. ; 4:3, s. 269-284
-
Tidskriftsartikel (refereegranskat)abstract
- Dalton is a powerful general-purpose program system for the study of molecular electronic structure at the Hartree-Fock, Kohn-Sham, multiconfigurational self-consistent-field, MOller-Plesset, configuration-interaction, and coupled-cluster levels of theory. Apart from the total energy, a wide variety of molecular properties may be calculated using these electronic-structure models. Molecular gradients and Hessians are available for geometry optimizations, molecular dynamics, and vibrational studies, whereas magnetic resonance and optical activity can be studied in a gauge-origin-invariant manner. Frequency-dependent molecular properties can be calculated using linear, quadratic, and cubic response theory. A large number of singlet and triplet perturbation operators are available for the study of one-, two-, and three-photon processes. Environmental effects may be included using various dielectric-medium and quantum-mechanics/molecular-mechanics models. Large molecules may be studied using linear-scaling and massively parallel algorithms. Dalton is distributed at no cost from for a number of UNIX platforms.
|
|
| 10. |
- Almberg, Wah-Sui, et al.
(författare)
-
An active agent portfolio algorithm
- 2003. - 1
-
Ingår i: Artificial intelligence and computer science. - : Nova publishers. - 1 59454 411 5 ; , s. 123-134
-
Bokkapitel (refereegranskat)abstract
- An algorithm for managing a portfolio of stocks using a trading agent is presented. A simulation game inspired by history-based Parrondo games is described. A performance measure is defined, with which various strategy mixes can be judged. Even when transaction costs are taken into account, active portfolio management (as opposed to Buy and Hold) is shown to be profitable.
|
|
