Zajem in analiza kontekstnih podatkov o uporabniku

Vrbančič, David

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Zajem in analiza kontekstnih podatkov o uporabniku
Authors:	ID Vrbančič, David (Author) ID Šumak, Boštjan (Mentor) More about this mentor...
Files:	MAG_Vrbancic_David_2018.pdf (3,02 MB) MD5: 0704AF1D40A6D46967C9B214385AC1B6 PID: 20.500.12556/dkum/b28bacd2-3264-4dea-ac23-982289d55e19
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	FERI - Faculty of Electrical Engineering and Computer Science
Abstract:	Svetovni splet je zaradi velikega števila podatkov postal bogat vir informacij, zato je pomembno, da uporabniku zagotovimo ustrezne informacije ob pravem času. Zajem in analiza kontekstnih podatkov nam lahko pomagata pri zagotavljanju ustreznih informacij. V sklopu magistrskega dela smo predstavili kje in na kakšen način pridobiti kontekstne podatke o uporabniku, kako le-te primerno shraniti, obdelati in predstaviti uporabniku. Na podlagi sistematičnega pregleda literature smo ugotovili, da je zavedanje uporabnikov o pridobivanju kontekstnih informacij odvisno od predstavitve uporabe kontekstnih podatkov (pravilniki o zasebnosti). Ali pridobivanje kontekstnih podatkov uporabnike moti, je odvisno od namena uporabe kontekstnih informacij. Starost je faktor, ki ima največji vpliv na zavedanje pridobivanja kontekstnih informacij med starejšimi in mlajšimi uporabniki. Za namen izboljšanja uporabniške izkušnje uporabniki niso pripravljeni deliti kontekstne informacije, vendar verjamejo, da lahko dostop do kontekstnih podatkov izboljša učinkovitost storitev. Na podlagi pridobljenih rezultatov testiranja klasifikatorja, ki temelji na naivnem Bayesovem algoritmu, smo ugotovili, da le-ta doseže visoko natančnost klasifikacije in je posledično primeren za klasifikacijo vsebine pridobljene iz socialnih omrežij. Možnosti za nadaljnje delo vidimo v nadgradnji inteligentnega sistema ter s tem izboljšanju natančnosti klasifikacije algoritma. Nadgrajen klasifikacijski algoritem bi pri klasifikaciji znal povezati in upoštevati čim več odvisnih dejavnikov, ki vplivajo na rezultat klasifikacije.
Keywords:	kontekst, kontekstni podatki (informacije), socialna omrežja, analiza podatkov, klasifikacija podatkov, inteligentni sistemi, sistemi priporočil
Place of publishing:	Maribor
Publisher:	[D. Vrbančič]
Year of publishing:	2018
PID:	20.500.12556/DKUM-70562
UDC:	004.777(043.2)
COBISS.SI-ID:	21508374
NUK URN:	URN:SI:UM:DK:P6DXQAIH
Publication date in DKUM:	21.06.2018
Views:	1613
Downloads:	168
Metadata:
Categories:	KTFMB - FERI
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:	The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:	24.05.2018

Secondary language

Language:	English
Title:	Capturing and analysing contextual information about the user
Abstract:	Due to a large amount of data, the World Wide Web has become a rich source of information, so it is important to provide relevant information to the user at the right time. Capturing and analysing contextual information can help us provide relevant information. As a part of master's thesis, we presented where and in what way to acquire contextual information about the user, and how to properly store, process and introduce it to the user. Based on a systematic review of the literature, we found that user awareness of contextual information acquisition depends on the presentation of the use of contextual information (privacy policies). A disturbance by contextual information acquisition among users depends on the purpose of using contextual information. Age is a factor with the greatest impact on the context-awareness among older and younger users. Users are not ready to share contextual information to improve user experience; however, they believe that access to contextual information can improve the efficiency of services. On the basis of the obtained classifier test results (using Naive Bayes algorithm), we found that the classifier achieves a high-precision classification and is thus suitable for the classification of social media content. We see the possibilities for further work in an upgrade of the intelligent system, thereby improving the algorithm precision classification. The upgraded classification algorithm would be able to connect and take into account as many dependent factors that affect the classification result.
Keywords:	context, contextual data (information), social media, data analysis, data classification, intelligent systems, recommendation systems

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back