Distance Prediction for Commercial Serial Crime Cases Using Time Delay Neural Networks

Ghazvini, Anahita; Abdullah, Siti Norul Huda Sheikh; Abdullah, Mohammed Ariff; Junoh, Nawawi; Kasim, Zainal Abidin

doi:10.14569/ijacsa.2015.061101

Cited by 1 publication

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ii. Point-of-Interest (POI) as shown in Figure 2( Representation of the spatial data is carried out to enable machine learning Kahaki 2011;Ghazvini 2015). The relationship between crime data and spatial feature attributes is defined as the distance between the crime location of interest and the spatial features.…”

Section: Methodsmentioning

confidence: 99%

Additional Feet-on-the-Street Deployment Method for Indexed Crime Prevention Initiative

Abdullah

Nordin

2018

View full text Add to dashboard Cite

Under the National Key Result Area (NKRA) Safe City Program's (SCP) Safe City Monitoring System (SCMS) initiative, the Royal Malaysian Police (RMP) manages the deployment of feet-on-the-street via the indexed crime hotspots. Working on an approach known as the Repeat Location Finder (RLF), the RMP determines the displacement of indexed crime on the hotspots and may deploy feet-on-the-streets at the identified displacement areas as crime prevention measures. This paper introduces another deployment capability by shifting the focus from the hotspots to the identified serial suspects. Displacement models work on the concentration of crime incidents and the propensity location where the concentration might shift to the surrounding immediate hotspots. This additional method on the other hand, works on the identified suspects and identifies the next location where the suspects might surface, which may take place beyond the distance and boundaries of the hotspots. The objective of this paper is to identify the spatial features that positively contribute towards this new method. The solutions to the objective have been tested on a dataset made available by the RMP comprising 74 serial criminal suspects around the areas of Selangor, Kuala Lumpur and Putrajaya, spanning from Jan 1 st to Dec 31 st 2013. The identification capability moves as high as 92.86%. The RMP has been presented with the results of this paper and it was concluded that this method may be applicable as another capability in managing the deployment of feet-on-the-street resources. ABSTRAK Melalui Sistem Pemantauan Bandar Selamat (SCMS) di bawah Program Bandar Selamat (SCP) Bidang Keberhasilan Utama Negara (NKRA), Polis DiRaja Malaysia (RMP) menguruskan pengerahan rondaan kaki berpandukan tempattempat jenayah berindeks yang berketumpatan tinggi. Melalui pendekatan yang dikenali sebagai Pencarian Lokasi Berulang (RLF), RMP menentukan anjakan jenayah berindeks dari tempat-tempat ketumpatan jenayah tinggi dan mengerahkan rondaan kaki ke kawasan-kawasan anjakan yang dikenalpasti untuk tujuan pencegahan jenayah. Makalah ini memperkenalkan satu lagi keupayaan pengerahan dengan memindahkan tumpuan daripada kawasan berketumpatan jenayah berindeks tinggi kepada suspek bersiri yang dikenalpasti. Model RLF mencari lokasi ketumpatan insiden jenayah berindeks dan melakarkan lokasi kecenderungan ketumpatan jenayah beranjak mengelilingi lokasi terdekat ketumpatan asal, manakala kaedah tambahan ini memberi tumpuan kepada suspek yang dikenalpasti dan mengenalpasti anjakan lokasi seterusnya di mana suspek mungkin muncul, di mana lokasinya mungkin berada di luar jarak dan sempadan lokasi ketumpatan. Objektif makalah ini adalah untuk mengenalpasti fitur spasial yang menyumbang kepada kaedah baharu ini. Penyelesaian kepada objektif ini telah diuji ke atas set data jenayah berindeks yang disediakan oleh RMP yang terdiri daripada 74 suspek jenayah indeks bersiri di sekitar kawasan Selangor, Kuala Lumpur dan Putrajaya bermula dari 1 Jan hingga 31 Dis 2013. Puncak kemampuan mengen...

show abstract

Section: Methodsmentioning

confidence: 99%

Additional Feet-on-the-Street Deployment Method for Indexed Crime Prevention Initiative

Abdullah

Nordin

2018

View full text Add to dashboard Cite

show abstract

Distance Prediction for Commercial Serial Crime Cases Using Time Delay Neural Networks

Cited by 1 publication

References 18 publications

Additional Feet-on-the-Street Deployment Method for Indexed Crime Prevention Initiative

Additional Feet-on-the-Street Deployment Method for Indexed Crime Prevention Initiative

Contact Info

Product

Resources

About