{"id":1391,"date":"2019-06-24T19:13:00","date_gmt":"2019-06-24T17:13:00","guid":{"rendered":"https:\/\/revista.newprojects.org\/?p=1391"},"modified":"2023-05-21T19:47:58","modified_gmt":"2023-05-21T17:47:58","slug":"analiza-masuratorilor-de-diagnosticare-a-diabetului-folosind-algoritmul-dbscan","status":"publish","type":"post","link":"https:\/\/revista.newprojects.org\/?p=1391","title":{"rendered":"Analiza m\u0103sur\u0103torilor de diagnosticare a diabetului folosind algoritmul DBSCAN"},"content":{"rendered":"\n<h2 class=\"wp-block-heading has-medium-font-size\"><strong>Implementare Python<\/strong><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><strong><em>Prof. T\u0103b\u00eerc\u0103 Angelica-Ioana, informatic\u0103 \u0219i TIC, Liceul Teoretic &#8222;Ion Heliade R\u0103dulescu&#8221; T\u00e2rgovi\u0219te<\/em><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong><em>Prof. T\u0103b\u00eerc\u0103 Nicolae-Radu<strong><em>, informatic\u0103 \u0219i TIC<\/em><\/strong>, Liceul Teoretic &#8222;Ion Heliade R\u0103dulescu&#8221; T\u00e2rgovi\u0219te<\/em><\/strong><\/p>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Contextul \u0219i importan\u021ba algoritmului DBSCAN<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">\u00cen cadrul domeniului clusteriz\u0103rii datelor, algoritmul DBSCAN (Density-Based Spatial Clustering of Applications with Noise) reprezint\u0103 o metod\u0103 de clusterizare bazat\u0103 pe densitate. Acest algoritm se bazeaz\u0103 pe ideea c\u0103 obiectele din aceea\u0219i regiune dens\u0103 a spa\u021biului de date ar trebui s\u0103 fie grupate \u00eempreun\u0103, \u00een timp ce regiunile rare sau punctele izolate sunt considerate zgomot \u0219i nu sunt incluse \u00een niciun cluster.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Schubert et al. (2017) au realizat o revizuire a algoritmului DBSCAN, subliniind importan\u021ba continu\u0103 a utiliz\u0103rii sale \u0219i oferind argumente pentru alegerea sa \u00een anumite situa\u021bii de clusterizare [1].<\/em> Importan\u021ba algoritmului DBSCAN const\u0103 \u00een capacitatea sa de a identifica clusterii de forme \u0219i dimensiuni arbitrar complexe, precum \u0219i capacitatea sa de a gestiona eficient datele care pot con\u021bine zgomot sau puncte izolate. Algoritmul DBSCAN a devenit o metod\u0103 popular\u0103 \u0219i eficient\u0103 \u00een analiza datelor \u0219i \u00een diverse domenii de aplicare, cum ar fi recunoa\u0219terea de tipare, analiza de imagini, analiza social\u0103, analiza traficului rutier, \u0219i multe altele.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Unul dintre scopurile cercet\u0103rii este de a explora \u0219i de a analiza algoritmul DBSCAN \u00een detaliu, \u00een vederea \u00een\u021belegerii sale teoretice \u0219i practice.<\/p>\n\n\n\n<!--more-->\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Principiile clusteriz\u0103rii bazate pe densitate<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Clusterizarea bazat\u0103 pe densitate este o abordare \u00een care clusterii sunt defini\u021bi ca regiuni dense de puncte de date separate de regiuni rar populate. Principalele principii ale clusteriz\u0103rii bazate pe densitate sunt [2]:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Nucleu (core): Este un punct de date \u00eenconjurat de un num\u0103r minim de puncte vecine, care indic\u0103 o regiune dens\u0103 a clusterului.<\/li>\n\n\n\n<li>Directly reachable points: Acestea sunt punctele de date situate \u00een interiorul sau \u00een apropierea regiunii dense \u0219i sunt considerate parte a clusterului.<\/li>\n\n\n\n<li>Reachability points: Acestea sunt punctele de date situate \u00een afara regiunii dense, dar pot fi conectate prin intermediul punctelor <em>Directly reachable points<\/em> \u0219i sunt considerate parte a clusterului.<\/li>\n\n\n\n<li>Puncte de zgomot (noise points): Acestea sunt punctele de date care nu pot fi atinse nici direct, nici indirect de nicio regiune dens\u0103 \u0219i nu sunt considerate parte a clusterului.<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Principiul de func\u021bionare al algoritmului DBSCAN Algoritmul DBSCAN <\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">(Density-Based Spatial Clustering of Applications with Noise) utilizeaz\u0103 principiile clusteriz\u0103rii bazate pe densitate pentru a identifica clusterii \u00eentr-un set de date. Principalele etape ale algoritmului sunt [1]:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Ini\u021bializarea: Se selecteaz\u0103 un punct de date nedescoperit aleator \u0219i se verific\u0103 vecin\u0103tatea sa pentru a determina dac\u0103 este un nucleu sau un punct de zgomot.<\/li>\n\n\n\n<li>Expandarea clusterului: Dac\u0103 un punct este un nucleu, se identific\u0103 toate punctele direct ating\u0103toare \u0219i se extinde clusterul prin ad\u0103ugarea acestor puncte \u0219i a punctelor <em>Reachability points<\/em> \u00een mod recursiv.<\/li>\n\n\n\n<li>Formarea clusterului: Cluster-ul se formeaz\u0103 prin ad\u0103ugarea continu\u0103 a punctelor <em>Directly reachable points<\/em> \u0219i a punctelor <em>Reachability points<\/em>.<\/li>\n\n\n\n<li>Identificarea punctelor de zgomot: Punctele care nu pot fi atinse nici direct, nici indirect de nicio regiune dens\u0103 sunt considerate puncte de zgomot \u0219i nu sunt atribuite niciunui cluster.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Algoritmul DBSCAN este robust la forme \u0219i dimensiuni variate ale clusterilor \u0219i poate detecta clusteri de orice form\u0103. De asemenea, poate identifica punctele de zgomot \u0219i nu necesit\u0103 specificarea num\u0103rului de clusteri \u00een avans.<a><\/a><\/p>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Descrierea algoritmului DBSCAN<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Algoritmul DBSCAN (Density-Based Spatial Clustering of Applications with Noise) este un algoritm de clusterizare bazat pe densitate. Acesta se bazeaz\u0103 pe ideea c\u0103 un cluster \u00eentr-un set de date este o regiune de densitate ridicat\u0103, separat\u0103 de regiuni de densitate sc\u0103zut\u0103 prin zone de zgomot.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Principiul de baz\u0103 al algoritmului DBSCAN const\u0103 \u00een identificarea punctelor de date &#8222;core&#8221; (nuclee) \u0219i a punctelor de date &#8222;border&#8221;. Un punct &#8222;core&#8221; este un punct de date care are cel pu\u021bin un num\u0103r minim predefinit de vecini \u00eentr-o anumit\u0103 raz\u0103 predefinit\u0103. Un punct &#8222;border&#8221; este un punct de date care are mai pu\u021bini vecini dec\u00e2t num\u0103rul minim, dar se afl\u0103 \u00een raza unui punct &#8222;core&#8221;. Punctele care nu sunt nici &#8222;core&#8221; nici &#8222;border&#8221; sunt considerate puncte de zgomot \u0219i nu sunt atribuite niciunui cluster [3].<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Procesul de clusterizare \u00een algoritmul DBSCAN \u00eencepe prin alegerea unui punct de date nevizitat \u0219i verificarea dac\u0103 acesta este un punct &#8222;core&#8221;. Dac\u0103 este, se formeaz\u0103 un nou cluster \u0219i se adaug\u0103 toate punctele din vecin\u0103tatea sa care satisfac condi\u021biile de &#8222;core&#8221; sau &#8222;border&#8221;. Acest proces continu\u0103 recursiv p\u00e2n\u0103 c\u00e2nd nu mai exist\u0103 puncte nevizitate care s\u0103 fie &#8222;core&#8221; sau s\u0103 fie \u00een vecin\u0103tatea punctelor &#8222;core&#8221;.<\/p>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Parametrii algoritmului \u0219i impactul lor asupra rezultatelor<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Algoritmul DBSCAN are doi parametri principali: raza (epsilon) \u0219i num\u0103rul minim de vecini (MinPts). Ace\u0219ti parametri influen\u021beaz\u0103 rezultatele clusteriz\u0103rii \u0219i trebuie ale\u0219i cu grij\u0103 \u00een func\u021bie de caracteristicile setului de date.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Raza (epsilon) determin\u0103 distan\u021ba maxim\u0103 p\u00e2n\u0103 la care un punct poate fi considerat vecin cu alt punct. O valoare mare a razei va duce la formarea de clusteri mai mari \u0219i la absorbirea unor clusteri mai mici \u00eentr-unul mai mare. Pe de alt\u0103 parte, o valoare mic\u0103 a razei poate duce la formarea de clusteri separa\u021bi care ar putea fi parte din acela\u0219i cluster.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Num\u0103rul minim de vecini (MinPts) specific\u0103 num\u0103rul minim de vecini pe care un punct trebuie s\u0103 \u00eei aib\u0103 \u00een raza specificat\u0103 pentru a fi considerat &#8222;core&#8221;. Aceasta controleaz\u0103 densitatea necesar\u0103 pentru a forma un cluster. Un MinPts mai mare va duce la formarea de clusteri mai restr\u00e2n\u0219i \u0219i va ignora punctele de zgomot, \u00een timp ce un MinPts mai mic va permite formarea de clusteri mai mari \u0219i poate include puncte de zgomot \u00een clusteri [1].<\/p>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">M\u0103suri de evaluare a calit\u0103\u021bii clusteriz\u0103rii<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Pentru a evalua performan\u021ba clusteriz\u0103rii realizate de algoritmul DBSCAN, exist\u0103 mai multe m\u0103suri \u0219i metode disponibile. Dou\u0103 dintre cele mai comune exemple sunt:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Indicele Silhouette<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Indicele Silhouette este o m\u0103sur\u0103 a calit\u0103\u021bii clusteriz\u0103rii, care ia \u00een considerare gradul de similaritate \u00eentre obiectele din acela\u0219i cluster \u0219i gradul de diferen\u021b\u0103 \u00eentre obiectele din clusterul respectiv \u0219i ceilal\u021bi clusteri. Valoarea indicelui Silhouette variaz\u0103 \u00eentre -1 \u0219i 1, unde o valoare mai mare indic\u0103 o clusterizare mai bun\u0103 [4].<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Validarea extern\u0103<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Validarea extern\u0103 implic\u0103 compararea rezultatelor algoritmului DBSCAN cu un set de date de referin\u021b\u0103, care con\u021bine etichetele reale ale obiectelor. Acesta poate implica utilizarea unor m\u0103suri precum precizia, acurate\u021bea sau m\u0103suri specifice cum ar fi Rand Index sau Fowlkes-Mallows Index. Validarea extern\u0103 ofer\u0103 o perspectiv\u0103 asupra c\u00e2t de bine se potrive\u0219te clusterizarea algoritmului DBSCAN cu datele de referin\u021b\u0103 [4].<\/p>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Compararea performan\u021bei DBSCAN cu al\u021bi algoritmi de clusterizare<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Pentru a compara performan\u021ba algoritmului DBSCAN cu al\u021bi algoritmi de clusterizare, se pot utiliza m\u0103suri de evaluare precum cele men\u021bionate anterior sau se pot realiza experimente comparative pe diverse seturi de date. Al\u021bi algoritmi de clusterizare populari, care pot fi lua\u021bi \u00een considerare \u00een compara\u021bie cu DBSCAN, sunt K-Means, Hierarchical Clustering sau Mean Shift.<\/p>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Avantajele \u0219i limit\u0103rile algoritmului DBSCAN [5] [6]<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Avantajele algoritmului DBSCAN:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Capacitatea de a identifica clusteri de forme \u0219i dimensiuni arbitrare: DBSCAN poate detecta clusteri de forme complexe \u0219i poate gestiona seturi de date \u00een care clusterii pot fi de dimensiuni \u0219i densit\u0103\u021bi variate. Nu necesit\u0103 s\u0103 se specifice forma sau dimensiunea clusterilor \u00een prealabil.<\/li>\n\n\n\n<li>Capacitatea de a identifica puncte de zgomot: Algoritmul poate identifica \u0219i clasifica punctele de date care nu apar\u021bin niciunui cluster ca puncte de zgomot. Acest lucru este util \u00een identificarea \u0219i eliminarea datelor nerelevante sau anomalii din seturile de date.<\/li>\n\n\n\n<li>Rezisten\u021ba la valori atipice: DBSCAN este rezistent la prezen\u021ba valorilor atipice \u00een setul de date. Datorit\u0103 utiliz\u0103rii densit\u0103\u021bii drept criteriu de formare a clusterilor, punctele atipice sau izolate nu vor fi incluse \u00eentr-un cluster, ci vor fi tratate ca puncte de zgomot.<\/li>\n\n\n\n<li>Scalabilitate: Algoritmul DBSCAN poate fi eficient implementat \u0219i aplicat pe seturi de date mari. Performan\u021ba sa nu scade semnificativ odat\u0103 cu cre\u0219terea dimensiunii setului de date.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Limit\u0103rile algoritmului DBSCAN:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Sensibilitate la parametri: Algoritmul DBSCAN necesit\u0103 specificarea a doi parametri cheie: raza de vecin\u0103tate \u0219i num\u0103rul minim de puncte pentru a forma un cluster. Alegerea incorect\u0103 a acestor parametri poate duce la rezultate inadecvate. Adesea, este necesar\u0103 o analiz\u0103 \u0219i ajustare manual\u0103 a parametrilor pentru a ob\u021bine rezultate optime.<\/li>\n\n\n\n<li>Sensibilitate la densitatea datelor: DBSCAN poate avea dificult\u0103\u021bi \u00een identificarea clusterilor \u00een seturile de date cu densit\u0103\u021bi inegale. Dac\u0103 exist\u0103 varia\u021bii semnificative ale densit\u0103\u021bii \u00een setul de date, pot ap\u0103rea dificult\u0103\u021bi \u00een detectarea clusterilor cu densit\u0103\u021bi sc\u0103zute sau \u00een separarea clusterilor densi de punctele de zgomot.<\/li>\n\n\n\n<li>Necesitatea de a alege o metric\u0103 adecvat\u0103: Algoritmul DBSCAN se bazeaz\u0103 pe m\u0103surarea distan\u021bei sau similarit\u0103\u021bii \u00eentre puncte. Alegerea unei metrici adecvate pentru setul de date \u0219i domeniul specific poate influen\u021ba rezultatele algoritmului.<\/li>\n\n\n\n<li>Necesitatea de a preprocesa datele: \u00cenainte de aplicarea algoritmului DBSCAN, este adesea necesar\u0103 o etap\u0103 de preprocesare a datelor, cum ar fi normalizarea sau scalarea, pentru a asigura coeren\u021ba valorilor \u0219i pentru a evita influen\u021ba dispropor\u021bionat\u0103 a anumitor caracteristici asupra rezultatelor clusteriz\u0103rii.<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading has-medium-font-size\">Rezultate<\/h2>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Setul de date \u0219i prelucrarea acestuia<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Setul de date utilizat \u00een acest studiu provine de la Institutul Na\u021bional al Diabetului \u0219i al Bolilor Digestive \u0219i Renale, care face parte din Institutul Na\u021bional de S\u0103n\u0103tate al Statelor Unite. Acesta este disponibil pe platforma Kaggle la urm\u0103toarea adres\u0103: <a href=\"https:\/\/www.kaggle.com\/datasets\/mathchi\/diabetes-data-set\" target=\"_blank\" rel=\"noopener\" title=\"\">https:\/\/www.kaggle.com\/datasets\/mathchi\/diabetes-data-set<\/a> .<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Acest set de date este alc\u0103tuit din m\u0103sur\u0103tori privind diabetul pentru pacientele de gen feminin, \u00een v\u00e2rst\u0103 de cel pu\u021bin 21 de ani, de origine indian\u0103. Scopul studiului este de a analiza gruparea acestor paciente utiliz\u00e2nd algoritmi de clusterizare, \u00een func\u021bie de indicatorii specifici diabetului. Setul de date con\u021bine urm\u0103toarele variabile:<\/p>\n\n\n\n<ol class=\"wp-block-list\" type=\"1\">\n<li><strong>ID<\/strong>: id-ul unic al fiec\u0103rei paciente;<\/li>\n\n\n\n<li><strong>Pregnancies<\/strong>: Num\u0103rul de sarcini;<\/li>\n\n\n\n<li><strong>Glucose<\/strong>: Concentra\u021bia de glucoz\u0103 plasmatic\u0103 la 2 ore \u00een cadrul unui test de toleran\u021b\u0103 la glucoz\u0103 oral\u0103;<\/li>\n\n\n\n<li><strong>BloodPressure<\/strong>: Tensiunea arterial\u0103 diastolic\u0103 (mm Hg);<\/li>\n\n\n\n<li><strong>SkinThickness<\/strong>: Grosimea pliului cutanat al tricepsului (mm);<\/li>\n\n\n\n<li><strong>Insulin<\/strong>: Insulin\u0103 seric\u0103 la 2 ore (mu U\/ml);<\/li>\n\n\n\n<li><strong>BMI<\/strong>: Indicele de mas\u0103 corporal\u0103 (greutate \u00een kg\/(\u00een\u0103l\u021bime \u00een m)^2);<\/li>\n\n\n\n<li><strong>DiabetesPedigreeFunction<\/strong>: Func\u021bia pedigree pentru diabet;<\/li>\n\n\n\n<li><strong>Age<\/strong>: v\u00e2rsta \u00een ani.<\/li>\n<\/ol>\n\n\n\n<p class=\"wp-block-paragraph\">\u00cenainte de a aplica algoritmii de clusterizare, setul de date a fost prelucrat pentru a asigura calitatea \u0219i consisten\u021ba acestuia. Acest proces de prelucrare a implicat urm\u0103torii pa\u0219i:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Cur\u0103\u021barea datelor<\/strong>: Identificarea \u0219i eliminarea eventualelor erori sau anomalii, precum valorile lips\u0103 sau incorecte.<\/li>\n\n\n\n<li><strong>Transformarea datelor<\/strong>: Conversia datelor \u00eentr-un format adecvat pentru aplicarea algoritmilor de clusterizare, cum ar fi normalizarea sau scalarea variabilelor.<\/li>\n\n\n\n<li><strong>Reducerea dimensiunilor<\/strong>: Selec\u021bionarea celor mai relevante variabile pentru analiza clusteriz\u0103rii, elimin\u00e2nd variabilele redundante sau coliniare, prin tehnici precum analiza componentelor principale (PCA) sau selec\u021bia univariat\u0103 a caracteristicilor.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Dup\u0103 prelucrarea \u0219i preg\u0103tirea setului de date, acesta a fost folosit pentru aplicarea \u0219i evaluarea algoritmului de clusterizare DBSCAN. Algoritmul a fost implementat \u00een limbajul de programare Python, folosindu-se biblioteci \u0219i pachete software specifice, cum ar fi scikit-learn \u0219i NumPy.<\/p>\n\n\n\n<h3 class=\"wp-block-heading has-medium-font-size\">Rezultatele ob\u021binute prin DBSCAN \u00een Python<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Analiza clusterilor prin histograme<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Prezentarea clusterilor<\/strong><\/p>\n\n\n\n<figure class=\"wp-block-image size-full is-resized\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN1.jpg\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN1.jpg\" alt=\"\" class=\"wp-image-1392\" width=\"841\" height=\"550\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN1.jpg 512w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN1-300x196.jpg 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN1-150x98.jpg 150w\" sizes=\"auto, (max-width: 841px) 100vw, 841px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 1. Cod surs\u0103 1 &#8211; algoritm Python DBSCAN<\/p>\n\n\n\n<figure class=\"wp-block-image size-full is-resized\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN2.jpg\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN2.jpg\" alt=\"\" class=\"wp-image-1393\" width=\"837\" height=\"478\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN2.jpg 515w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN2-300x171.jpg 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN2-150x86.jpg 150w\" sizes=\"auto, (max-width: 837px) 100vw, 837px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 2. Cod surs\u0103 2 &#8211; algoritm Python DBSCAN<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>V\u00e2rsta \u2013 <\/strong>pacien\u021bii care au v\u00e2rsta cea mai mic\u0103 se situeaz\u0103 \u00een clusterul 1.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN3.png\"><img loading=\"lazy\" decoding=\"async\" width=\"505\" height=\"331\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN3.png\" alt=\"\" class=\"wp-image-1394\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN3.png 505w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN3-300x197.png 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN3-150x98.png 150w\" sizes=\"auto, (max-width: 505px) 100vw, 505px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 3. Histogram\u0103 DBSCAN (id = v\u00e2rst\u0103)<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Presiunea arterial\u0103 <\/strong>&#8211; pacien\u021bii care au presiunea arterial\u0103 cea mai mare se situeaz\u0103 \u00een clusterul 1.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN4.png\"><img loading=\"lazy\" decoding=\"async\" width=\"501\" height=\"331\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN4.png\" alt=\"\" class=\"wp-image-1395\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN4.png 501w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN4-300x198.png 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN4-150x99.png 150w\" sizes=\"auto, (max-width: 501px) 100vw, 501px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 4. Histogram\u0103 DBSCAN (id = presiune arterial\u0103)<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Indicele de mas\u0103 corporal\u0103 <\/strong>&#8211; pacien\u021bii care au indicele de mas\u0103 corporal\u0103 cel mai mare se situeaz\u0103 \u00een clusterul 1.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN5.png\"><img loading=\"lazy\" decoding=\"async\" width=\"505\" height=\"331\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN5.png\" alt=\"\" class=\"wp-image-1396\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN5.png 505w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN5-300x197.png 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN5-150x98.png 150w\" sizes=\"auto, (max-width: 505px) 100vw, 505px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 5. Histogram\u0103 DBSCAN (id = indice de mas\u0103 corporal\u0103)<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Probabilitatea de diabet pe baza antecedentelor familiale &#8211; <\/strong>pacien\u021bii care au probabilitatea de diabet cea mai mare se situeaz\u0103 \u00een clusterul 1.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN6.png\"><img loading=\"lazy\" decoding=\"async\" width=\"505\" height=\"331\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN6.png\" alt=\"\" class=\"wp-image-1397\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN6.png 505w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN6-300x197.png 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN6-150x98.png 150w\" sizes=\"auto, (max-width: 505px) 100vw, 505px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 6. Histogram\u0103 DBSCAN (id = pb. de diabet pe baza antecedentelor familiale)<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Cantitatea de glucoz\u0103 &#8211; <\/strong>pacien\u021bii care au cantitatea de glucoz\u0103 cea mai mare se situeaz\u0103 \u00een&nbsp; clusterul 1.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN7.png\"><img loading=\"lazy\" decoding=\"async\" width=\"505\" height=\"331\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN7.png\" alt=\"\" class=\"wp-image-1398\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN7.png 505w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN7-300x197.png 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN7-150x98.png 150w\" sizes=\"auto, (max-width: 505px) 100vw, 505px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 7. Histogram\u0103 DBSCAN (id = cantitatea de glucoz\u0103)<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Insulin\u0103 &#8211; <\/strong>pacien\u021bii care au insulina cea mai mic\u0103 se situeaz\u0103 \u00een clusterul 1.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN8.png\"><img loading=\"lazy\" decoding=\"async\" width=\"505\" height=\"331\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN8.png\" alt=\"\" class=\"wp-image-1399\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN8.png 505w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN8-300x197.png 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN8-150x98.png 150w\" sizes=\"auto, (max-width: 505px) 100vw, 505px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 8. Histogram\u0103 DBSCAN (id = insulin\u0103)<strong><br><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Num\u0103rul de na\u0219teri &#8211; <\/strong>pacien\u021bii care au num\u0103rul de na\u0219teri cel mai mare se situeaz\u0103 \u00een clusterul 1.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN9.png\"><img loading=\"lazy\" decoding=\"async\" width=\"505\" height=\"331\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN9.png\" alt=\"\" class=\"wp-image-1400\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN9.png 505w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN9-300x197.png 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN9-150x98.png 150w\" sizes=\"auto, (max-width: 505px) 100vw, 505px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 9. Histogram\u0103 DBSCAN (id = nr. de na\u0219teri)<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Grosimea pielii &#8211; <\/strong>pacien\u021bii care au grosimea pielii cea mai mare se situeaz\u0103 \u00een clusterul 1.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN10.png\"><img loading=\"lazy\" decoding=\"async\" width=\"501\" height=\"331\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN10.png\" alt=\"\" class=\"wp-image-1401\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN10.png 501w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN10-300x198.png 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN10-150x99.png 150w\" sizes=\"auto, (max-width: 501px) 100vw, 501px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 10. Histogram\u0103 DBSCAN (id = grosimea pielii)<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Graficul parti\u021biei eviden\u021biaz\u0103 formarea a nou\u0103 clusteri, dintre care doar c1 av\u00e2nd datele u\u0219or mai grupate.<\/strong><\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN11.jpg\"><img loading=\"lazy\" decoding=\"async\" width=\"514\" height=\"328\" src=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN11.jpg\" alt=\"\" class=\"wp-image-1402\" srcset=\"https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN11.jpg 514w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN11-300x191.jpg 300w, https:\/\/revista.newprojects.org\/wp-content\/uploads\/DBSCAN11-150x96.jpg 150w\" sizes=\"auto, (max-width: 514px) 100vw, 514px\" \/><\/a><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Figur\u0103 11. Graficul parti\u021biei DBSCAN<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u00cen concluzie, DBSCAN (Implementare Python): Aplicarea algoritmului DBSCAN a condus la formarea a nou\u0103 clusteri, dintre care doar c1 av\u00e2nd datele u\u0219or mai grupate. De exemplu, pacien\u021bii care au v\u00e2rsta cea mai mic\u0103, presiunea arterial\u0103 cea mai mare, indicele de mas\u0103 corporal\u0103 cel mai mare, probabilitatea de diabet cea mai mare, cantitatea de glucoz\u0103 cea mai mare, insulina cea mai mic\u0103, num\u0103rul de na\u0219teri cel mai mare, \u0219i grosimea pielii cea mai mare se situeaz\u0103 \u00een clusterul 1.<\/p>\n\n\n\n<h2 class=\"wp-block-heading has-medium-font-size\">Referin\u021be bibliografice<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>[1]: <\/strong><a href=\"https:\/\/dl.acm.org\/doi\/10.1145\/3068335\">https:\/\/dl.acm.org\/doi\/10.1145\/3068335<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Schubert, E., Sander, J., Ester, M., Kriegel, H. P., &amp; Xu, X. (2017). DBSCAN revisited, revisited: why and how you should (still) use DBSCAN. ACM Transactions on Database Systems (TODS), 42(3), 19<\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>[2]: <\/strong><a href=\"http:\/\/www.cs.ubbcluj.ro\/~lauras\/test\/docs\/school\/MIRPR\/lectures\/05_ML.pdf\">http:\/\/www.cs.ubbcluj.ro\/~lauras\/test\/docs\/school\/MIRPR\/lectures\/05_ML.pdf<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>[3]: <\/strong><a href=\"https:\/\/towardsdatascience.com\/dbscan-make-density-based-clusters-by-hand-2689dc335120\">https:\/\/towardsdatascience.com\/dbscan-make-density-based-clusters-by-hand-2689dc335120<\/a><strong><\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>[4]: <\/strong><a href=\"https:\/\/link.springer.com\/article\/10.1007\/BF01908075\">https:\/\/link.springer.com\/article\/10.1007\/BF01908075<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Rousseeuw, P.J. (1987). Silhouettes: A Graphical Aid to the Interpretation and Validation of Cluster Analysis. <\/em><em><\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>[5]: <\/strong><a href=\"https:\/\/dl.acm.org\/doi\/10.5555\/3001460.3001507\">https:\/\/dl.acm.org\/doi\/10.5555\/3001460.3001507<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Ester, M., Kriegel, H.P., Sander, J., &amp; Xu, X. (1996). A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. In Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining (KDD-96).<\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>[6]:<\/strong> <a href=\"https:\/\/ieeexplore.ieee.org\/document\/914876\">https:\/\/ieeexplore.ieee.org\/document\/914876<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Jin, W., Tung, A.K.H., Han, J., &amp; Wang, W. (2001). Mining Top-k Covering Rule Groups for Gene Expression Data. In Proceedings of the 17th International Conference on Data Engineering (ICDE 2001).<\/em><em><\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Implementare Python Prof. T\u0103b\u00eerc\u0103 Angelica-Ioana, informatic\u0103 \u0219i TIC, Liceul Teoretic &#8222;Ion Heliade R\u0103dulescu&#8221; T\u00e2rgovi\u0219te Prof. T\u0103b\u00eerc\u0103 Nicolae-Radu, informatic\u0103 \u0219i TIC, Liceul Teoretic &#8222;Ion Heliade R\u0103dulescu&#8221; T\u00e2rgovi\u0219te Contextul \u0219i importan\u021ba algoritmului DBSCAN \u00cen cadrul domeniului clusteriz\u0103rii datelor, algoritmul DBSCAN (Density-Based Spatial Clustering of Applications with Noise) reprezint\u0103 o metod\u0103 de clusterizare bazat\u0103 pe densitate. Acest algoritm&#8230;<\/p>\n","protected":false},"author":7,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-1391","post","type-post","status-publish","format-standard","hentry","category-fara-categorie"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=\/wp\/v2\/posts\/1391","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1391"}],"version-history":[{"count":9,"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=\/wp\/v2\/posts\/1391\/revisions"}],"predecessor-version":[{"id":1419,"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=\/wp\/v2\/posts\/1391\/revisions\/1419"}],"wp:attachment":[{"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1391"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1391"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/revista.newprojects.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1391"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}