| 1. |
- Ansell, Ricky, et al.
(författare)
-
Contamination monitoring in the forensic DNA laboratory and a simple graphical model for unbiased EPG classification
- 2011
-
Ingår i: Book of Abstracts. ; , s. 199-
-
Konferensbidrag (refereegranskat)abstract
- In this work we present a procedure for contamination monitoring in a trace search and recovery area and agraphical classification model. The recent launch of more sensitive and robust amplification kits increases thepossibility to detect minute amounts of trace DNA. As a consequence this enhances our need to establish eliminationdatabases and demands for an increased awareness on how to avoid contamination. DNA contaminatingthe evidence somewhere along the forensic process has the potential to destroy the evidence or totally confuseand mislead the crime investigation. In the forensic laboratory specific areas are designated for different partsof the process: trace search and recovery, pre-PCR, post-PCR etc. Work procedures and cleaning routines areadapted to minimise the risk of contamination. Monitoring presence of DNA in the laboratory environment, onspecific surfaces or instruments of interest, is one way to assess these risks and will in addition increase ourknowledge on how to improve cleaning routines and behaviour in the lab. A monitoring process needs to someextent be standardised in order to become unbiased and independent on an individual level, regarding bothwhere and how samples are taken and how the results are classified. The graphical model constitutes a lineartransformation of a three-dimensional “credit system” based on alleles, markers and peak heights, into a twodimensional classification. The standardisation allows results to be compared over time, and if applied to otherwork-areas comparison between different parts of the process will be possible.
|
|
| 2. |
- Digréus, P., et al.
(författare)
-
Contamination monitoring in the forensic DNA laboratory and a simple graphical model for unbiased EPG classification
- 2011
-
Ingår i: Forensic Science International: Genetics, Supplement Series. - : Elsevier. - 1875-1768. ; 3:1, s. e299-e300
-
Tidskriftsartikel (refereegranskat)abstract
- Monitoring presence and level of background DNA in forensic DNA laboratory environments can be used to control work routines and cleaning procedures and to follow changes in these, as well as being an indicator for increased/decreased contamination risk. Previous monitoring routines as sampling and interpretation have not been standardised, making it difficult to compare between different sampling events and observe potential trends. Factor analysis was used to generate a simple graphical classification model for unbiased ranking of electropherograms, which can be modified according to user's need, taking into account number of detected alleles, markers and peak height.
|
|
| 3. |
- Garcia, Johan, 1970-, et al.
(författare)
-
Towards Video Flow Classification at a Million Encrypted Flows Per Second
- 2018
-
Ingår i: Proceedings of 32nd International Conference on Advanced Information Networking and Applications (AINA). - Krakow : IEEE. - 9781538621967 - 9781538621950
-
Konferensbidrag (refereegranskat)abstract
- As end-to-end encryption on the Internet is becoming more prevalent, techniques such as deep packet inspection (DPI) can no longer be expected to be able to classify traffic. In many cellular networks a large fraction of all traffic is video traffic, and being able to divide flows in the network into video and non-video can provide considerable traffic engineering benefits. In this study we examine machine learning based flow classification using features that are available also for encrypted flows. Using a data set of several several billion packets from a live cellular network we examine the obtainable classification performance for two different ensemble-based classifiers. Further, we contrast the classification performance of a statistical-based feature set with a less computationally demanding alternate feature set. To also examine the runtime aspects of the problem, we export the trained models and use a tailor-made C implementation to evaluate the runtime performance. The results quantify the trade-off between classification and runtime performance, and show that up to 1 million classifications per second can be achieved for a single core. Considering that only the subset of flows reaching some minimum flow length will need to be classified, the results are promising with regards to deployment also in scenarios with very high flow arrival rates.
|
|
