Anomaly Detection for Temporal Data using Long Short-Term ...

DEGREE PROJECT IN INFORMATION AND COMMUNICATION

TECHNOLOGY,

SECOND CYCLE, 30 CREDITS

STOCKHOLM, SWEDEN 2017

Anomaly Detection for Temporal

Data using Long Short-Term

Memory (LSTM)

AKASH SINGH

KTH ROYAL INSTITUTE OF TECHNOLOGY

SCHOOL OF INFORMATION AND COMMUNICATION TECHNOLOGY

Anomaly Detection for Temporal Data using Long

Short-Term Memory (LSTM)

AKASH SINGH

Master¡¯s Thesis at KTH Information and Communication Technology

Supervisor: Daniel Gillblad

Examiner: Magnus Boman

Industrial Supervisors: Mona Matti, Rickard C?ster (Ericsson)

TRITA-ICT-EX-2017:124

Abstract

We explore the use of Long short-term memory (LSTM)

for anomaly detection in temporal data. Due to the challenges in obtaining labeled anomaly datasets, an unsupervised approach is employed. We train recurrent neural

networks (RNNs) with LSTM units to learn the normal

time series patterns and predict future values. The resulting prediction errors are modeled to give anomaly scores.

We investigate different ways of maintaining LSTM state,

and the effect of using a fixed number of time steps on

LSTM prediction and detection performance. LSTMs are

also compared to feed-forward neural networks with fixed

size time windows over inputs. Our experiments, with three

real-world datasets, show that while LSTM RNNs are suitable for general purpose time series modeling and anomaly

detection, maintaining LSTM state is crucial for getting desired results. Moreover, LSTMs may not be required at all

for simple time series.

Keywords: LSTM; RNN; anomaly detection; time series;

deep learning

Abstrakt

Vi unders?ker Long short-term memory (LSTM) f?r avvikelsedetektion i tidsseriedata. P? grund av sv?righeterna

i att hitta data med etiketter s? har ett o?vervakat angreppss?tt anv?nts. Vi tr?nar rekursiva neuronn?t (RNN)

med LSTM-noder f?r att l?ra modellen det normala tidsseriem?nstret och prediktera framtida v?rden. Vi unders?ker olika s?tt av att beh?lla LSTM-tillst?ndet och effekter av att anv?nda ett konstant antal tidssteg p? LSTMprediktionen och avvikelsedetektionsprestandan. LSTM ?r

ocks? j?mf?rda med vanliga neuronn?t med fasta tidsf?nster ?ver indata. V?ra experiment med tre verkliga dataset

visar att ?ven om LSTM RNN ?r till?mpbara f?r generell

tidsseriemodellering och avvikelsedetektion s? ?r det avg?rande att beh?lla LSTM-tillst?ndet f?r att f? de ?nskade

resultaten. Dessutom ?r det inte n?dv?ndigt att anv?nda

LSTM f?r enkla tidsserier.

Keywords: LSTM; RNN; avvikelsedetektion; tidsserier; djupt

l?rande

Acknowledgements

I am grateful to Magnus Boman for his time, feedback, and genuine kindness. His

guidance during times of struggle was essential in completing this thesis. I would

like to thank my supervisor, Daniel Gillblad, for his inputs and ideas for the thesis

as well as the future. My warmest regards to Mona Matti and Rickard C?ster for

giving me the opportunity to do this thesis at Ericsson and their support. I would

also like to extend my gratitude to the entire Machine Intelligence team at Ericsson,

Kista, for welcoming me to the team, and their engagement during the project. I

would like to appreciate the contribution of my opponents, Staffan Aldenfalk and

Andrea Azzini, for their critique of my work. Special thanks to my family and

friends for their continuous encouragement and motivation. Finally, I would like

to thank my wife, Deepta, without whose love and support none of this would be

possible.

Tack!

Thank You!

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download