Malmö University Publications
Change search
Link to record
Permanent link

Direct link
Publications (6 of 6) Show all publications
Khoshkangini, R., Tajgardan, M., Mashhadi, P., Rognvaldsson, T. & Tegnered, D. (2024). Optimal Task Grouping Approach in Multitask Learning. In: Luo, B Wu, ZG Cheng, C Li, H Li, C (Ed.), Neural Information Processing: 30th International Conference, ICONIP 2023, Changsha, China, November 20–23, 2023, Proceedings, Part VI. Paper presented at 30th International Conference on Neural Information Processing (ICONIP) of the Asia-Pacific-Neural-Network-Society (APNNS), NOV 20-23, 2023, Changsha, PEOPLES R CHINA (pp. 206-225). Springer, 14452
Open this publication in new window or tab >>Optimal Task Grouping Approach in Multitask Learning
Show others...
2024 (English)In: Neural Information Processing: 30th International Conference, ICONIP 2023, Changsha, China, November 20–23, 2023, Proceedings, Part VI / [ed] Luo, B Wu, ZG Cheng, C Li, H Li, C, Springer, 2024, Vol. 14452, p. 206-225Conference paper, Published paper (Refereed)
Abstract [en]

Multi-task learning has become a powerful solution in which multiple tasks are trained together to leverage the knowledge learned from one task to improve the performance of the other tasks. However, the tasks are not always constructive on each other in the multi-task formulation and might play negatively during the training process leading to poor results. Thus, this study focuses on finding the optimal group of tasks that should be trained together for multi-task learning in an automotive context. We proposed a multi-task learning approach to model multiple vehicle long-term behaviors using low-resolution data and utilized gradient descent to efficiently discover the optimal group of tasks/vehicle behaviors that can increase the performance of the predictive models in a single training process. In this study, we also quantified the contribution of individual tasks in their groups and to the other groups' performance. The experimental evaluation of the data collected from thousands of heavy-duty trucks shows that the proposed approach is promising.

Place, publisher, year, edition, pages
Springer, 2024
Series
Lecture Notes in Computer Science, ISSN 0302-9743, E-ISSN 1611-3349 ; 14452
Keywords
Machine Learning, Vehicle Usage Behavior, Multitask learning
National Category
Computer Systems
Identifiers
urn:nbn:se:mau:diva-66154 (URN)10.1007/978-981-99-8076-5_15 (DOI)001148055700015 ()978-981-99-8075-8 (ISBN)978-981-99-8076-5 (ISBN)
Conference
30th International Conference on Neural Information Processing (ICONIP) of the Asia-Pacific-Neural-Network-Society (APNNS), NOV 20-23, 2023, Changsha, PEOPLES R CHINA
Available from: 2024-02-27 Created: 2024-02-27 Last updated: 2024-02-27Bibliographically approved
Khoshkangini, R., Tajgardan, M., Lundström, J., Rabbani, M. & Tegnered, D. (2023). A Snapshot-Stacked Ensemble and Optimization Approach for Vehicle Breakdown Prediction. Sensors, 23(12), Article ID 5621.
Open this publication in new window or tab >>A Snapshot-Stacked Ensemble and Optimization Approach for Vehicle Breakdown Prediction
Show others...
2023 (English)In: Sensors, E-ISSN 1424-8220, Vol. 23, no 12, article id 5621Article in journal (Refereed) Published
Abstract [en]

Predicting breakdowns is becoming one of the main goals for vehicle manufacturers so as to better allocate resources, and to reduce costs and safety issues. At the core of the utilization of vehicle sensors is the fact that early detection of anomalies facilitates the prediction of potential breakdown issues, which, if otherwise undetected, could lead to breakdowns and warranty claims. However, the making of such predictions is too complex a challenge to solve using simple predictive models. The strength of heuristic optimization techniques in solving np-hard problems, and the recent success of ensemble approaches to various modeling problems, motivated us to investigate a hybrid optimization- and ensemble-based approach to tackle the complex task. In this study, we propose a snapshot-stacked ensemble deep neural network (SSED) approach to predict vehicle claims (in this study, we refer to a claim as being a breakdown or a fault) by considering vehicle operational life records. The approach includes three main modules: Data pre-processing, Dimensionality Reduction, and Ensemble Learning. The first module is developed to run a set of practices to integrate various sources of data, extract hidden information and segment the data into different time windows. In the second module, the most informative measurements to represent vehicle usage are selected through an adapted heuristic optimization approach. Finally, in the last module, the ensemble machine learning approach utilizes the selected measurements to map the vehicle usage to the breakdowns for the prediction. The proposed approach integrates, and uses, the following two sources of data, collected from thousands of heavy-duty trucks: Logged Vehicle Data (LVD) and Warranty Claim Data (WCD). The experimental results confirm the proposed system's effectiveness in predicting vehicle breakdowns. By adapting the optimization and snapshot-stacked ensemble deep networks, we demonstrate how sensor data, in the form of vehicle usage history, contributes to claim predictions. The experimental evaluation of the system on other application domains also indicated the generality of the proposed approach.

Place, publisher, year, edition, pages
MDPI, 2023
Keywords
breakdown prediction, optimization, deep neural networks, ensemble learning
National Category
Vehicle Engineering
Identifiers
urn:nbn:se:mau:diva-61954 (URN)10.3390/s23125621 (DOI)001015804000001 ()37420787 (PubMedID)2-s2.0-85163933766 (Scopus ID)
Available from: 2023-08-17 Created: 2023-08-17 Last updated: 2023-08-22Bibliographically approved
Khoshkangini, R., Mashhadi, P., Tegnered, D., Lundström, J. & Rögnvaldsson, T. (2023). Predicting Vehicle Behavior Using Multi-task Ensemble Learning. Expert systems with applications, 212, 118716-118716, Article ID 118716.
Open this publication in new window or tab >>Predicting Vehicle Behavior Using Multi-task Ensemble Learning
Show others...
2023 (English)In: Expert systems with applications, ISSN 0957-4174, E-ISSN 1873-6793, Vol. 212, p. 118716-118716, article id 118716Article in journal (Refereed) Published
Abstract [en]

Vehicle utilization analysis is an essential tool for manufacturers to understand customer needs, improve equipment uptime, and to collect information for future vehicle and service development. Typically today, this behavioral modeling is done on high-resolution time-resolved data with features such as GPS position and fuel consumption. However, high-resolution data is costly to transfer and sensitive from a privacy perspective. Therefore, such data is typically only collected when the customer pays for extra services relying on that data. This motivated us to develop a multi-task ensemble approach to transfer knowledge from the high-resolution data and enable vehicle behavior prediction from low-resolution but high dimensional data that is aggregated over time in the vehicles. This study proposes a multi-task snapshot-stacked ensemble (MTSSE) deep neural network for vehicle behavior prediction by considering vehicles’ low-resolution operational life records. The multi-task ensemble approach utilizes the measurements to map the low-frequency vehicle usage to the vehicle behaviors defined from the high-resolution time-resolved data. Two data sources are integrated and used: high-resolution data called Dynafleet, and low-resolution so-called Logged Vehicle Data (LVD). The experimental results demonstrate the proposed approach’s effectiveness in predicting the vehicle behavior from low frequency data. With the suggested multi-task snapshot-stacked ensemble deep network, it is shown how low-resolution sensor data can highly contribute to predicting multiple vehicle behaviors simultaneously while using only one single training process.

Place, publisher, year, edition, pages
Elsevier, 2023
National Category
Information Systems
Identifiers
urn:nbn:se:mau:diva-55080 (URN)10.1016/j.eswa.2022.118716 (DOI)000870841300003 ()2-s2.0-85138456634 (Scopus ID)
Available from: 2022-09-22 Created: 2022-09-22 Last updated: 2023-07-05Bibliographically approved
Jamali, M., Davidsson, P., Khoshkangini, R., Ljungqvist, M. G. & Mihailescu, R.-C. (2023). Specialized Indoor and Outdoor Scene-specific Object Detection Models. In: Osten, Wolfgang (Ed.), Sixteenth International Conference on Machine Vision (ICMV 2023): . Paper presented at International Conference on Machine Vision (ICMV 2023), Nov. 15-18, 2023, Yerevan, Armenia.
Open this publication in new window or tab >>Specialized Indoor and Outdoor Scene-specific Object Detection Models
Show others...
2023 (English)In: Sixteenth International Conference on Machine Vision (ICMV 2023) / [ed] Osten, Wolfgang, 2023Conference paper, Published paper (Refereed)
Abstract [en]

Object detection is a critical task in computer vision with applications across various domains, ranging from autonomous driving to surveillance systems. Despite extensive research on improving the performance of object detection systems, identifying all objects in different places remains a challenge. The traditional object detection approaches focus primarily on extracting and analyzing visual features without considering the contextual information about the places of objects. However, entities in many real-world scenarios closely relate to their surrounding environment, providing crucial contextual cues for accurate detection. This study investigates the importance and impact of places of images (indoor and outdoor) on object detection accuracy. To this purpose, we propose an approach that first categorizes images into two distinct categories: indoor and outdoor. We then train and evaluate three object detection models (indoor, outdoor, and general models) based on YOLOv5 and 19 classes of the PASCAL VOC dataset and 79 classes of COCO dataset that consider places. The experimental evaluations show that the specialized indoor and outdoor models have higher mAP (mean Average Precision) to detect objects in specific environments compared to the general model that detects objects found both indoors and outdoors. Indeed, the network can detect objects more accurately in similar places with common characteristics due to semantic relationships between objects and their surroundings, and the network’s misdetection is diminished. All the results were analyzed statistically with t-tests.

Series
Proceedings of SPIE, ISSN 0277-786X, E-ISSN 1996-756X ; 13072
Keywords
object detection, YOLOv5, indoor object detection, outdoor object detection, scene classification
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:mau:diva-66441 (URN)10.1117/12.3023479 (DOI)001208308300024 ()2-s2.0-85191658757 (Scopus ID)9781510674622 (ISBN)9781510674639 (ISBN)
Conference
International Conference on Machine Vision (ICMV 2023), Nov. 15-18, 2023, Yerevan, Armenia
Available from: 2024-03-22 Created: 2024-03-22 Last updated: 2024-05-20Bibliographically approved
Zolfaghari, M., Fadishei, H., Tajgardan, M. & Khoshkangini, R. (2022). Stock Market Prediction Using Multi-Objective Optimization. In: 2022 12th International Conference on Computer and Knowledge Engineering (ICCKE): . Paper presented at 2022 12th International Conference on Computer and Knowledge Engineering (ICCKE), 17-18 November 2022, Mashhad, Islamic Republic of Iran. Institute of Electrical and Electronics Engineers (IEEE)
Open this publication in new window or tab >>Stock Market Prediction Using Multi-Objective Optimization
2022 (English)In: 2022 12th International Conference on Computer and Knowledge Engineering (ICCKE), Institute of Electrical and Electronics Engineers (IEEE), 2022Conference paper, Published paper (Refereed)
Abstract [en]

Forecasting in financial markets is challenging due to the inherent randomness of financial data sources and the vast number of factors that affect the market trends. Thus, it is essential to find informative elements within the vast number of available factors to enhance the performance of the predictive models in such a vital context. This makes the feature selection process an integral part of the financial prediction. In this paper, we propose a multi-objective evolutionary algorithm to reduce the number of features employed to predict the yearly performance of the US stock market. The primary idea is to select a smaller set of features with the slightest similarity and the best prediction accuracy. In this practice, we have utilized genetic algorithm, XGBoost and correlation in order to obtain a more diverse set of features which increases the performance. Experiential results show that our proposed approach is able to reduce the number of features significantly while maintaining comparable prediction accuracy.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2022
Series
Proceedings of the Internatinal eConference on Computer and Knowledge Engineering, ISSN 2375-1304, E-ISSN 2643-279X
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:mau:diva-56498 (URN)10.1109/iccke57176.2022.9960002 (DOI)2-s2.0-85143767437 (Scopus ID)978-1-6654-7613-3 (ISBN)978-1-6654-7614-0 (ISBN)
Conference
2022 12th International Conference on Computer and Knowledge Engineering (ICCKE), 17-18 November 2022, Mashhad, Islamic Republic of Iran
Available from: 2022-12-07 Created: 2022-12-07 Last updated: 2024-02-05Bibliographically approved
Khoshkangini, R., Rani Kalia, N., Ashwathanarayana, S., Orand, A., Maktobian, J. & Tajgardan, M. (2022). Vehicle Usage Extraction Using Unsupervised Ensemble Approach. In: Proceedings of SAI Intelligent Systems Conference: . Paper presented at Intelligent Systems and Applications, 1-2 September 2022, Amsterdam, The Netherlands (pp. 588-604). Springer
Open this publication in new window or tab >>Vehicle Usage Extraction Using Unsupervised Ensemble Approach
Show others...
2022 (English)In: Proceedings of SAI Intelligent Systems Conference, Springer, 2022, p. 588-604Conference paper, Published paper (Refereed)
Abstract [en]

Current heavy vehicles are equipped with hundreds of sensors that are used to continuously collect data in motion. The logged data enables researchers and industries to address three main transportation issues related to performance (e.g. fuel consumption, breakdown), environment (e.g., emission reduction), and safety (e.g. reducing vehicle accidents and incidents during maintenance activities). While according to the American Transportation Research Institute (ATRI), the operational cost of heavy vehicles is around 59%59% of overall costs, there are limited studies demonstrating the specific impacts of external factors (e.g. weather and road conditions, driver behavior) on vehicle performance. In this work, vehicle usage modeling was studied based on time to determine the different usage styles of vehicles and how they can affect vehicle performance. An ensemble clustering approach was developed to extract vehicle usage patterns and vehicle performance taking into consideration logged vehicle data (LVD) over time. Analysis results showed a strong correlation between driver behavior and vehicle performance that would require further investigation.

Place, publisher, year, edition, pages
Springer, 2022
Series
Lecture Notes in Networks and Systems, ISSN 2367-3370, E-ISSN 2367-3389 ; 542
National Category
Other Engineering and Technologies not elsewhere specified
Identifiers
urn:nbn:se:mau:diva-54745 (URN)10.1007/978-3-031-16072-1_43 (DOI)000890312800043 ()2-s2.0-85137975588 (Scopus ID)978-3-031-16071-4 (ISBN)978-3-031-16072-1 (ISBN)
Conference
Intelligent Systems and Applications, 1-2 September 2022, Amsterdam, The Netherlands
Available from: 2022-09-06 Created: 2022-09-06 Last updated: 2024-02-05Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-3797-4605

Search in DiVA

Show all publications