{"id":54058,"date":"2019-12-17T00:00:00","date_gmt":"2019-12-16T21:00:00","guid":{"rendered":"https:\/\/prohoster.info\/blog\/blog_prohoster\/14-open-source-proektov-dlya-prokachki-data-science-masterstva-easy-normal-hard"},"modified":"2020-02-18T14:02:02","modified_gmt":"2020-02-18T11:02:02","slug":"14-open-source-proektov-dlya-prokachki-data-science-masterstva-easy-normal-hard","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/novosti-interneta\/14-open-source-proektov-dlya-prokachki-data-science-masterstva-easy-normal-hard","title":{"rendered":"14 projekte me burim t\u00eb hapur p\u00ebr t\u00eb p\u00ebrmir\u00ebsuar aft\u00ebsit\u00eb n\u00eb Data Science (leht\u00eb, normal, e v\u00ebshtir\u00eb)","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Data Science p\u00ebr fillestar\u00ebt<\/i><\/p>\n

1. Analiza e Ndjenjave (Analiza e ndjenjave p\u00ebrmes tekstit)<\/h3>\n

\n\"14
\n
\nShikoni implementimin e plot\u00eb t\u00eb projektit Data Science me p\u00ebrdorimin e kodit burim \u2014 Projekti i Analiz\u00ebs s\u00eb Ndjenjave n\u00eb R<\/a><\/noindex>.<\/p>\n

Sentiment Analysis \u2014 \u044d\u0442\u043e \u0430\u043d\u0430\u043b\u0438\u0437 \u0441\u043b\u043e\u0432 \u0434\u043b\u044f \u043e\u043f\u0440\u0435\u0434\u0435\u043b\u0435\u043d\u0438\u044f \u043d\u0430\u0441\u0442\u0440\u043e\u0435\u043d\u0438\u0439 \u0438 \u043c\u043d\u0435\u043d\u0438\u0439, \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u043c\u043e\u0433\u0443\u0442 \u0431\u044b\u0442\u044c \u043f\u043e\u043b\u043e\u0436\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u043c\u0438 \u0438\u043b\u0438 \u043e\u0442\u0440\u0438\u0446\u0430\u0442\u0435\u043b\u044c\u043d\u044b\u043c\u0438. \u042d\u0442\u043e \u0442\u0438\u043f \u043a\u043b\u0430\u0441\u0441\u0438\u0444\u0438\u043a\u0430\u0446\u0438\u0438, \u043f\u0440\u0438 \u043a\u043e\u0442\u043e\u0440\u043e\u043c \u043a\u043b\u0430\u0441\u0441\u044b \u043c\u043e\u0433\u0443\u0442 \u0431\u044b\u0442\u044c \u0434\u0432\u043e\u0438\u0447\u043d\u044b\u043c\u0438 (\u043f\u043e\u043b\u043e\u0436\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u043c\u0438 \u0438 \u043e\u0442\u0440\u0438\u0446\u0430\u0442\u0435\u043b\u044c\u043d\u044b\u043c\u0438) \u0438\u043b\u0438 \u043c\u043d\u043e\u0436\u0435\u0441\u0442\u0432\u0435\u043d\u043d\u044b\u043c\u0438 (\u0441\u0447\u0430\u0441\u0442\u043b\u0438\u0432\u044b\u043c\u0438, \u0437\u043b\u044b\u043c\u0438, \u0433\u0440\u0443\u0441\u0442\u043d\u044b\u043c\u0438, \u043f\u0440\u043e\u0442\u0438\u0432\u043d\u044b\u043c\u0438 …). \u041c\u044b \u0440\u0435\u0430\u043b\u0438\u0437\u0443\u0435\u043c \u044d\u0442\u043e\u0442 Data Science \u043f\u0440\u043e\u0435\u043a\u0442 \u043d\u0430 \u044f\u0437\u044b\u043a\u0435 R \u0438 \u0431\u0443\u0434\u0435\u043c \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u044c \u043d\u0430\u0431\u043e\u0440 \u0434\u0430\u043d\u043d\u044b\u0445 \u0432 \u043f\u0430\u043a\u0435\u0442\u0435 \u00abjaneaustenR\u00bb. \u041c\u044b \u0431\u0443\u0434\u0435\u043c \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u044c \u0441\u043b\u043e\u0432\u0430\u0440\u0438 \u043e\u0431\u0449\u0435\u0433\u043e \u043d\u0430\u0437\u043d\u0430\u0447\u0435\u043d\u0438\u044f, \u0442\u0430\u043a\u0438\u0435 \u043a\u0430\u043a AFINN, bing \u0438 loughran, \u0432\u044b\u043f\u043e\u043b\u043d\u044f\u0442\u044c \u0432\u043d\u0443\u0442\u0440\u0435\u043d\u043d\u0435\u0435 \u0441\u043e\u0435\u0434\u0438\u043d\u0435\u043d\u0438\u0435, \u0438 \u0432 \u043a\u043e\u043d\u0446\u0435 \u043c\u044b \u0441\u043e\u0437\u0434\u0430\u0434\u0438\u043c \u043e\u0431\u043b\u0430\u043a\u043e \u0441\u043b\u043e\u0432, \u0447\u0442\u043e\u0431\u044b \u043e\u0442\u043e\u0431\u0440\u0430\u0437\u0438\u0442\u044c \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442.<\/p>\n

Gjuha:<\/b> R
\nGrupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> janeaustenR
\n<\/a><\/noindex><\/p>\n

\"14<\/a><\/noindex>

\nArtikulli \u00ebsht\u00eb p\u00ebrkthyer me mb\u00ebshtetje nga kompania EDISON Software, e cila krijon dhoma provimi virtuale p\u00ebr dyqanet me shum\u00eb marka<\/a><\/noindex>, si dhe teston softuerin<\/a><\/noindex>.<\/p><\/blockquote>\n

2. Zbulimi i Fake News (Zbulimi i lajmeve false)<\/h3>\n

\n\u00c7 elevoni aft\u00ebsit\u00eb tuaja n\u00eb nj\u00eb nivel t\u00eb ri duke punuar n\u00eb nj\u00eb projekt p\u00ebr Data Science p\u00ebr fillestar\u00ebt \u2014 identifikimi i lajmeve t\u00eb rreme me Python<\/a><\/noindex>.<\/p>\n

\"14
\n
\nLajmet e rreme jan\u00eb informacion i rrem\u00eb q\u00eb p\u00ebrhapet p\u00ebrmes mediave sociale dhe burimeve t\u00eb tjera p\u00ebr arritjen e q\u00ebllimeve politike. N\u00eb k\u00ebt\u00eb ide projekti p\u00ebr Data Science, ne do t\u00eb p\u00ebrdorim Python p\u00ebr t\u00eb nd\u00ebrtuar nj\u00eb model q\u00eb mund t\u00eb p\u00ebrcaktoj\u00eb sakt\u00ebsisht n\u00ebse nj\u00eb lajm \u00ebsht\u00eb i v\u00ebrtet\u00eb apo i rrem\u00eb. Ne do t\u00eb krijojm\u00eb TfidfVectorizer dhe do t\u00eb p\u00ebrdorim PassiveAggressiveClassifier p\u00ebr t\u00eb klasifikuar lajmet si \"t\u00eb v\u00ebrteta\" ose \"t\u00eb rreme\". Ne do t\u00eb p\u00ebrdorim nj\u00eb grup t\u00eb dh\u00ebnash me form\u00ebn 7796 \u00d7 4 dhe do t\u00eb realizojm\u00eb gjith\u00e7ka n\u00eb Jupyter Lab.<\/p>\n

Gjuha:<\/b> Python<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> news.csv<\/p>\n

3. Identifikimi i S\u00ebmundjes s\u00eb Parkinsonit<\/h3>\n

\nShkoni p\u00ebrpara duke punuar n\u00eb iden\u00eb e projektit Data Science - identifikimi i s\u00ebmundjes s\u00eb Parkinsonit me XGBoost<\/a><\/noindex>.<\/p>\n

\"14
\n
\nKemi filluar t\u00eb p\u00ebrdorim Shkenc\u00ebn e T\u00eb Dh\u00ebnave p\u00ebr t\u00eb p\u00ebrmir\u00ebsuar sh\u00ebndet\u00ebsin\u00eb dhe sh\u00ebrbimet \u2014 n\u00ebse mund t\u00eb parashikojm\u00eb s\u00ebmundjen n\u00eb fazat e para, do t\u00eb kemi shum\u00eb p\u00ebrfitime. Prandaj, n\u00eb k\u00ebt\u00eb ide projekti p\u00ebr Shkenc\u00ebn e T\u00eb Dh\u00ebnave do t\u00eb m\u00ebsojm\u00eb si t\u00eb identifikojm\u00eb s\u00ebmundjen e Parkinsonit duke p\u00ebrdorur Python. Kjo \u00ebsht\u00eb nj\u00eb s\u00ebmundje neurodegjenerative, progresive e sistemit nervor qendror, e cila ndikon n\u00eb l\u00ebvizje dhe shkakton dridhje dhe ngurt\u00ebsi. Ajo ndikon te neuronet q\u00eb prodhojn\u00eb dopamin\u00eb n\u00eb tru, dhe \u00e7do vit, ajo prek m\u00eb shum\u00eb se 1 milion njer\u00ebz n\u00eb Indi.<\/p>\n

Gjuha:<\/b> Python<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> UCI ML Parkinsons dataset<\/p>\n

Projekte t\u00eb Shkenc\u00ebs s\u00eb T\u00eb Dh\u00ebnave me v\u00ebshtir\u00ebsi t\u00eb mesme<\/i><\/p>\n

4. Njohja e Emocioneve n\u00eb Fjal\u00eb (Speech Emotion Recognition)<\/h3>\n

\nShikoni realizimin e plot\u00eb t\u00eb shembujve t\u00eb projekteve t\u00eb Shkenc\u00ebs s\u00eb T\u00eb Dh\u00ebnave \u2014 njohja e fjal\u00ebs me Librosa<\/a><\/noindex>.<\/p>\n

\"14
\n
\nTani le t\u00eb m\u00ebsojm\u00eb t\u00eb p\u00ebrdorim biblioteka t\u00eb ndryshme. Ky projekt n\u00eb Data Science p\u00ebrdor librosa p\u00ebr njohjen e z\u00ebrit. SER\u2014a \u00ebsht\u00eb procesi i identifikimit t\u00eb emocioneve njer\u00ebzore dhe gjendjeve affective p\u00ebrmes z\u00ebrit. Duke qen\u00eb se ne p\u00ebrdorim tonin dhe lart\u00ebsin\u00eb e tonit p\u00ebr t\u00eb shprehur emocionet me z\u00eb, SER \u00ebsht\u00eb n\u00eb raund. Por, pasi emocionet jan\u00eb subjektive, annotimi i tingujve \u00ebsht\u00eb nj\u00eb detyr\u00eb e komplikuar. Ne do t\u00eb p\u00ebrdorim funksionet mfcc, chroma dhe mel dhe do t\u00eb p\u00ebrdorim setin e t\u00eb dh\u00ebnave RAVDESS p\u00ebr njohjen e emocioneve. Ne do t\u00eb krijojm\u00eb nj\u00eb klasifikues MLPC p\u00ebr k\u00ebt\u00eb model.<\/p>\n

Gjuha:<\/b> Python<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> RAVDESS dataset<\/p>\n

5. Gender and Age Detection (Detektimi i Gjinis\u00eb dhe Mosh\u00ebs)<\/h3>\n

\nMerrni v\u00ebmendjen e pun\u00ebdh\u00ebn\u00ebsve me projektin m\u00eb t\u00eb ri n\u00eb Data Science \u2014 detektimi i gjinis\u00eb dhe mosh\u00ebs me OpenCV<\/a><\/noindex>.<\/p>\n

\"14
\n
\nKy \u00ebsht\u00eb nj\u00eb projekt interesant Data Science me Python. Duke p\u00ebrdorur vet\u00ebm nj\u00eb imazh, do t\u00eb m\u00ebsoni t\u00eb parashikoni gjinin\u00eb dhe mosh\u00ebn e nj\u00eb personi. N\u00eb k\u00ebt\u00eb projekt do t\u2019ju prezantojm\u00eb me Computer Vision dhe principet e tij. Ne do t\u00eb nd\u00ebrtojm\u00eb nj\u00eb rrjet nervor konvolucional<\/a><\/noindex> dhe do t\u00eb p\u00ebrdorim modelet e trajnuara nga Tal Hassner dhe Gil Levi p\u00ebr setin e t\u00eb dh\u00ebnave Adience. Gjat\u00eb rrug\u00ebs do t\u00eb p\u00ebrdorim disa skedar\u00eb .pb, .pbtxt, .prototxt dhe .caffemodel.<\/p>\n

Gjuha:<\/b> Python<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> Adience<\/p>\n

6. Uber Data Analysis (Analiza e t\u00eb Dh\u00ebnave t\u00eb Uberit)<\/h3>\n

\nShikoni realizimin e plot\u00eb t\u00eb projektit Data Science me kodin burimor \u2014 Projekti i Analiz\u00ebs s\u00eb Uber n\u00eb R<\/a><\/noindex>.<\/p>\n

\"14
\n
\nKy \u00ebsht\u00eb nj\u00eb projekt vizualizimi t\u00eb dh\u00ebnash me ggplot2, n\u00eb t\u00eb cilin do t\u00eb p\u00ebrdorim R dhe bibliotekat e tij p\u00ebr t\u00eb analizuar parametrat e ndrysh\u00ebm. Do t\u00eb p\u00ebrdorim setin e t\u00eb dh\u00ebnave Uber Pickups n\u00eb New York dhe do t\u00eb krijojm\u00eb vizualizime p\u00ebr periudha t\u00eb ndryshme t\u00eb vitit. Kjo na tregon se si koha ndikon n\u00eb udh\u00ebtimet e klient\u00ebve.<\/p>\n

Gjuha:<\/b> R<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> Dataset-i Uber Pickups n\u00eb qytetin e New York-ut<\/p>\n

7. Zbulimi i dueriz si rezultat i lodhjes (Driver Drowsiness detection)<\/h3>\n

\nP\u00ebrmir\u00ebsoni aft\u00ebsit\u00eb tuaja duke punuar n\u00eb nj\u00eb projekt t\u00eb lart\u00eb t\u00eb Data Science \u2014 nj\u00eb sistem zbulimi lodhjeje me OpenCV & Keras<\/a><\/noindex>.<\/p>\n

\"14
\n
\nT\u00eb drejtuarit n\u00eb gjum\u00eb \u00ebsht\u00eb jasht\u00ebzakonisht e rrezikshme, dhe \u00e7do vit ndodhin rreth nj\u00eb mij\u00eb aksidente p\u00ebr shkak se shofer\u00ebt bien n\u00eb gjum\u00eb gjat\u00eb drejtimit. N\u00eb k\u00ebt\u00eb projekt n\u00eb Python, ne do t\u00eb krijojm\u00eb nj\u00eb sistem q\u00eb mund t\u00eb zbuloj\u00eb shofer\u00eb t\u00eb lodhur dhe t\u2019i paralajm\u00ebroj\u00eb ata me nj\u00eb sinjal akustik.<\/p>\n

Ky projekt \u00ebsht\u00eb realizuar duke p\u00ebrdorur Keras dhe OpenCV. Ne do t\u00eb p\u00ebrdorim OpenCV p\u00ebr zbulimin e fytyr\u00ebs dhe syve, dhe me Keras ne do t\u00eb klasifikojm\u00eb gjendjen e syve (T\u00eb hapur ose T\u00eb mbyllur) duke p\u00ebrdorur metoda t\u00eb rrjeteve nervore t\u00eb thella.<\/p>\n

8. Chatbot<\/h3>\n

\nKrijoni nj\u00eb chatbot me Python dhe b\u00ebni nj\u00eb hap p\u00ebrpara n\u00eb karrier\u00ebn tuaj \u2014 Chatbot me NLTK & Keras<\/a><\/noindex>.<\/p>\n

\"14
\n
\nChatbot\u00ebt jan\u00eb nj\u00eb pjes\u00eb e pandashme e biznesit. Shum\u00eb nd\u00ebrmarrje duhet t\u00eb ofrojn\u00eb sh\u00ebrbime p\u00ebr klient\u00ebt e tyre, dhe p\u00ebr t\u2019i sh\u00ebrbyer atyre k\u00ebrkohet shum\u00eb pun\u00eb, koh\u00eb dhe p\u00ebrpjekje. Chatbot\u00ebt mund t\u00eb automatizojn\u00eb nj\u00eb pjes\u00eb t\u00eb madhe t\u00eb nd\u00ebrveprimit me klient\u00ebt, duke u p\u00ebrgjigjur n\u00eb disa pyetje t\u00eb shpeshta q\u00eb b\u00ebjn\u00eb klient\u00ebt. N\u00eb thelb, ka dy lloje chatbot-\u00ebsh: Domain-specific dhe Open-domain. Chatbot\u00ebt Domain-specific p\u00ebrdoren shpesh p\u00ebr t\u00eb zgjidhur nj\u00eb problem t\u00eb caktuar. K\u00ebshtu, ju duhet ta p\u00ebrshtatni at\u00eb p\u00ebr t\u00eb funksionuar n\u00eb m\u00ebnyr\u00eb efektive n\u00eb sektorin tuaj. Chatbot\u00ebt Open-domain mund t'u b\u00ebhen \u00e7do pyetje, prandaj k\u00ebrkojn\u00eb nj\u00eb mas\u00eb t\u00eb madhe t\u00eb dh\u00ebnash p\u00ebr t'u trajnuar.<\/p>\n

Grupi i t\u00eb dh\u00ebnave:<\/b> Skedari json p\u00ebr Intentet<\/p>\n

Gjuha:<\/b> Python<\/p>\n

Projekte t\u00eb avancuara t\u00eb Shkenc\u00ebs s\u00eb t\u00eb Dh\u00ebnave<\/i><\/p>\n

9. Gjeneratori i p\u00ebrshkrimeve t\u00eb imazheve<\/h3>\n

\nKontrolloni realizimin e plot\u00eb t\u00eb projektit me kodin burimor \u2014 Gjeneratori i p\u00ebrshkrimeve t\u00eb imazheve me CNN & LSTM<\/a><\/noindex>.<\/p>\n

\"14
\n
\nP\u00ebrshkrimi i asaj q\u00eb \u00ebsht\u00eb n\u00eb imazh \u00ebsht\u00eb nj\u00eb detyr\u00eb e leht\u00eb p\u00ebr njer\u00ebzit, por p\u00ebr kompjuter\u00ebt, imazhi \u00ebsht\u00eb thjesht nj\u00eb grup numrash q\u00eb p\u00ebrfaq\u00ebsojn\u00eb vler\u00ebn e ngjyr\u00ebs s\u00eb secilit pixel. Kjo \u00ebsht\u00eb nj\u00eb detyr\u00eb e v\u00ebshtir\u00eb p\u00ebr kompjuter\u00ebt. T\u00eb kuptosh at\u00eb q\u00eb ndodhet n\u00eb imazh dhe pastaj t\u00eb krijosh nj\u00eb p\u00ebrshkrim n\u00eb nj\u00eb gjuh\u00eb natyrale (p\u00ebr shembull, anglisht) \u00ebsht\u00eb nj\u00eb detyr\u00eb tjet\u00ebr e v\u00ebshtir\u00eb. Ky projekt p\u00ebrdor metoda t\u00eb m\u00ebsimit t\u00eb thell\u00eb, n\u00eb t\u00eb cilat ne implementojm\u00eb nj\u00eb Rrjet Neuronal Konvolucional (CNN) me nj\u00eb Rrjet Neuronal Rekurent (LSTM) p\u00ebr t\u00eb krijuar nj\u00eb gjenerator p\u00ebrshkrimesh p\u00ebr imazhe.<\/p>\n

Grupi i t\u00eb dh\u00ebnave:<\/b> Flickr 8K<\/p>\n

Gjuha:<\/b> Python<\/p>\n

Korniza:<\/b> Keras<\/p>\n

10. Zbulesa e Mashtrimit t\u00eb Kartave t\u00eb Kreditit<\/h3>\n

\nB\u00ebni maksimumin duke punuar n\u00eb iden\u00eb e projektit Data Science \u2014 zbulimi i mashtrimeve me karta krediti p\u00ebrmes m\u00ebsimit t\u00eb makinerive<\/a><\/noindex>.<\/p>\n

\"14
\n
\nDerisa t\u00eb arrini k\u00ebt\u00eb pik\u00eb, keni filluar t\u00eb kuptoni metodat dhe konceptet. Le t\u00eb kalojm\u00eb n\u00eb disa projekte m\u00eb t\u00eb avancuara n\u00eb fush\u00ebn e shkencave t\u00eb t\u00eb dh\u00ebnave. N\u00eb k\u00ebt\u00eb projekt do t\u00eb p\u00ebrdorim gjuh\u00ebn R me algoritme t\u00eb till\u00eb si pem\u00ebt e vendimmarrjes<\/a><\/noindex>, regresioni logjistik, rrjetet nervore t\u00eb artificiale dhe klasifikuesi i p\u00ebrmir\u00ebsimit t\u00eb gradientit. Ne do t\u00eb p\u00ebrdorim nj\u00eb set t\u00eb dh\u00ebnash t\u00eb transaksioneve me karta p\u00ebr t\u00eb klasifikuar transaksionet e kartave t\u00eb kreditit si mashtruese ose t\u00eb v\u00ebrteta. Ne do t\u00eb zgjidhim modele t\u00eb ndryshme dhe do t\u00eb nd\u00ebrtojm\u00eb kurba performanc\u00ebs.<\/p>\n

Gjuha:<\/b> R<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> Seti i t\u00eb dh\u00ebnave t\u00eb Transaksioneve me Karta<\/p>\n

11. Sistema i rekomandimeve p\u00ebr filma<\/h3>\n

\nEksploroni realizimin e projektit m\u00eb t\u00eb mir\u00eb t\u00eb Shkenc\u00ebs s\u00eb t\u00eb Dh\u00ebnave me Kodin Burimor \u2014 Sistema e rekomandimeve p\u00ebr filma n\u00eb R<\/a><\/noindex><\/p>\n

\"14
\n
\nN\u00eb k\u00ebt\u00eb projekt t\u00eb Shkenc\u00ebs s\u00eb t\u00eb Dh\u00ebnave, ne do t\u00eb p\u00ebrdorim R p\u00ebr t\u00eb realizuar rekomandimet p\u00ebr filma p\u00ebrmes m\u00ebsimit t\u00eb makinerive. Sistemi i rekomandimeve d\u00ebrgon sugjerime p\u00ebrdoruesve p\u00ebrmes nj\u00eb procesi filtrimi, duke u bazuar n\u00eb preferencat e p\u00ebrdoruesve t\u00eb tjer\u00eb dhe historin\u00eb e shikimeve. N\u00ebse A dhe B e p\u00eblqejn\u00eb Home Alone, dhe B e do Mean Girls, mund t'i sugjerohet A \u2014 mund t'i p\u00eblqej\u00eb edhe atij. Kjo lejon q\u00eb klient\u00ebt t\u00eb nd\u00ebrveprojn\u00eb me platform\u00ebn.<\/p>\n

Gjuha:<\/b> R<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> Seti i t\u00eb dh\u00ebnave MovieLens<\/p>\n

12. Segmente t\u00eb klient\u00ebve<\/h3>\n

\nB\u00ebni p\u00ebrshtypje t\u00eb pun\u00ebdh\u00ebn\u00ebsve me ndihm\u00ebn e projektit t\u00eb Shkenc\u00ebs s\u00eb t\u00eb Dh\u00ebnave (duke p\u00ebrfshir\u00eb kodin burimor) \u2014 Segmentimi i klient\u00ebve p\u00ebrmes m\u00ebsimit t\u00eb makinerive<\/a><\/noindex>.<\/p>\n

\"14
\n
\nSegmentimi i bler\u00ebsve \u00ebsht\u00eb nj\u00eb aplikacion popullor m\u00ebsimi i paushtuar (unsupervised learning)<\/a><\/noindex>. Duke p\u00ebrdorur grupimin, kompanit\u00eb p\u00ebrcaktojn\u00eb segmentet e klient\u00ebve p\u00ebr t\u00eb punuar me nj\u00eb baz\u00eb t\u00eb mundshme p\u00ebrdoruesish. Ato ndajn\u00eb klient\u00ebt n\u00eb grupe sipas karakteristikave t\u00eb p\u00ebrbashk\u00ebta, si\u00e7 jan\u00eb gjinia, mosha, interesat dhe zakonet e shpenzimit, n\u00eb m\u00ebnyr\u00eb q\u00eb t\u00eb mund t\u00eb shesin produktet e tyre n\u00eb m\u00ebnyr\u00eb efikase p\u00ebr \u00e7do grup. Ne do t\u00eb p\u00ebrdorim grupimin K-means<\/a><\/noindex>, si dhe t\u00eb visualizojm\u00eb shp\u00ebrndarjen sipas gjinis\u00eb dhe mosh\u00ebs. M\u00eb pas do t\u00eb analizojm\u00eb t\u00eb ardhurat e tyre vjetore dhe nivelin e shpenzimeve.<\/p>\n

Gjuha:<\/b> R<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> datasetin Mall_Customers<\/p>\n

13. Klassifikimi i Kancerit t\u00eb Gjirit<\/h3>\n

\nShikoni realizimin e plot\u00eb t\u00eb projektit t\u00eb Shkenc\u00ebs s\u00eb t\u00eb Dh\u00ebnave n\u00eb Python \u2014 Klassifikimi i Kancerit t\u00eb Gjirit duke p\u00ebrdorur t\u00eb M\u00ebsuarit e thell\u00eb<\/a><\/noindex>.<\/p>\n

\"14
\n
\nDuke u kthyer te kontributi mjek\u00ebsor i shkenc\u00ebs s\u00eb t\u00eb dh\u00ebnave, le t\u00eb m\u00ebsojm\u00eb t\u00eb identifikojm\u00eb kancerin e gjirit duke p\u00ebrdorur Python. Ne do t\u00eb p\u00ebrdorim setin e t\u00eb dh\u00ebnave IDC_regular p\u00ebr t\u00eb identifikuar karcinom\u00ebn invazive t\u00eb kanaleve, lloji m\u00eb i zakonsh\u00ebm i kancerit t\u00eb gjirit. Ai zhvillohet n\u00eb kanalet e qum\u00ebshtit, duke dep\u00ebrtuar n\u00eb indet fibrilare ose yndyrore t\u00eb gjirit nga jasht\u00eb kanalit. N\u00eb k\u00ebt\u00eb projekt shkencor t\u00eb mbledhjes s\u00eb t\u00eb dh\u00ebnave ne do t\u00eb p\u00ebrdorim Deep Learning<\/a><\/noindex> dhe bibliotek\u00ebn Keras p\u00ebr klasifikim.<\/p>\n

Gjuha:<\/b> Python<\/p>\n

Grupi i t\u00eb dh\u00ebnave\/Paketa:<\/b> IDC_regular<\/p>\n

14. Traffic Signs Recognition (NJohja e Shenjave Rrugore)<\/h3>\n

\nArritja e sakt\u00ebsis\u00eb n\u00eb teknologjin\u00eb e drejtimit t\u00eb automjeteve me projektin e shkenc\u00ebs s\u00eb t\u00eb dh\u00ebnave p\u00ebr njohjen e shenjave rrugore duke p\u00ebrdorur CNN<\/a><\/noindex> me kod t\u00eb hapur.<\/p>\n

\"14
\n
\nTabelat e trafikut dhe rregullat e qarkullimit jan\u00eb shum\u00eb t\u00eb r\u00ebnd\u00ebsishme p\u00ebr \u00e7do shofer p\u00ebr t\u00eb shmangur aksidentet. P\u00ebr t\u00eb respektuar rregullin, s\u00eb pari duhet t\u00eb kuptohet si duket nj\u00eb tabel\u00eb e trafikut. Nj\u00ebri duhet t\u00eb m\u00ebsoj\u00eb t\u00eb gjitha tabelat e trafikut para se t'i jepet leja p\u00ebr t\u00eb drejtuar ndonj\u00eb mjet. Por tani numri i mjeteve autonome \u00ebsht\u00eb n\u00eb rritje dhe n\u00eb t\u00eb ardhmen t\u00eb ardhmen, njeriu nuk do t\u00eb drejtoj\u00eb m\u00eb vet\u00eb makin\u00ebn. N\u00eb projektin \u201cNjohja e Tabelave t\u00eb Trafikut\u201d do t\u00eb m\u00ebsoni si programi mund t\u00eb njoh\u00eb llojin e tabelave t\u00eb trafikut duke marr\u00eb imazhin si hyrje. Grupi i t\u00eb dh\u00ebnave p\u00ebr njohjen e tabelave t\u00eb trafikut n\u00eb Gjermani (GTSRB) p\u00ebrdoret p\u00ebr nd\u00ebrtimin e nj\u00eb rrjeti nervor t\u00eb thell\u00eb p\u00ebr t\u00eb njohur klas\u00ebn n\u00eb t\u00eb cil\u00ebn p\u00ebrket tabela e trafikut. Ne gjithashtu po krijojm\u00eb nj\u00eb nd\u00ebrfaqe grafike t\u00eb thjesht\u00eb p\u00ebr nd\u00ebrveprimin me aplikacionin.<\/p>\n

Gjuha:<\/b> Python<\/p>\n

Grupi i t\u00eb dh\u00ebnave:<\/b> GTSRB (German Traffic Sign Recognition Benchmark)<\/p>\n

Lexoni m\u00eb shum\u00eb<\/h3>\n

<\/p>\n