. . . . . ", esperimentazioan, diagnostikoan eta informazioaren berreskurapenean doitasuna eta estaldura elementu multzo bat hautematean izandako eraginkortasunaren neurriak dira. Doitasuna hautemandako elementu guztietan zuzen hautemandako elementuen proportzioa da eta estaldura hauteman beharreko elementu guztietatik hautemandako elementuen proportzioa da. Adibidez, espezie bateko 20 aleetatik gaixo daudenak hauteman behar dira eta 8 ale dira gaixotzat jo direnak, horietatik 6 benetako gaixoak izanik; datu horietatik doitasuna 6/8=%75 da (hau da, hautemandako gaixo guztietatik %75 dira benetako gaixoak) eta estaldura 6/20=%30 da (hau da, gaixo guztietatik %30a hauteman da)."@eu . . . . "\u0412\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u0442\u0430 \u043F\u043E\u0432\u043D\u043E\u0442\u0430"@uk . . "\uC774\uC9C4 \uBD84\uB958 \uAE30\uBC95(binary classification)\uC744 \uC0AC\uC6A9\uD558\uB294 \uD328\uD134 \uC778\uC2DD\uACFC \uC815\uBCF4 \uAC80\uC0C9 \uBD84\uC57C\uC5D0\uC11C, \uC815\uBC00\uB3C4\uB294 \uAC80\uC0C9\uB41C \uACB0\uACFC\uB4E4 \uC911 \uAD00\uB828 \uC788\uB294 \uAC83\uC73C\uB85C \uBD84\uB958\uB41C \uACB0\uACFC\uBB3C\uC758 \uBE44\uC728\uC774\uACE0, \uC7AC\uD604\uC728\uC740 \uAD00\uB828 \uC788\uB294 \uAC83\uC73C\uB85C \uBD84\uB958\uB41C \uD56D\uBAA9\uB4E4 \uC911 \uC2E4\uC81C \uAC80\uC0C9\uB41C \uD56D\uBAA9\uB4E4\uC758 \uBE44\uC728\uC774\uB2E4. \uB530\uB77C\uC11C \uC815\uBC00\uB3C4\uC640 \uC7AC\uD604\uC728 \uBAA8\uB450 \uAD00\uB828\uB3C4(Relevance)\uC758 \uCE21\uC815 \uAE30\uC900 \uBC0F \uC9C0\uC2DD\uC744 \uD1A0\uB300\uB85C \uD558\uACE0 \uC788\uB2E4."@ko . . . . . . "\u062F\u0642\u0629 \u0648\u0645\u0631\u0627\u062C\u0639\u0629"@ar . . "\u0627\u0644\u062F\u0642\u0629 \u0648\u0627\u0644\u0625\u0631\u062C\u0627\u0639 \u0647\u0648 \u0645\u0641\u0647\u0648\u0645 \u0641\u064A \u0627\u0644\u0631\u064A\u0627\u0636\u064A\u0627\u062A \u0648\u0641\u064A \u0639\u0644\u0645 \u0627\u0644\u062D\u0627\u0633\u0648\u0628 \u0627\u0644\u0645\u062E\u062A\u0635 \u0641\u064A \u0645\u062C\u0627\u0644 \u0627\u0633\u062A\u0631\u062C\u0627\u0639 \u0627\u0644\u0645\u0639\u0644\u0648\u0645\u0627\u062A. \u0627\u0644\u062F\u0642\u0629 \u0647\u064A \u0645\u0639\u064A\u0627\u0631 \u064A\u0642\u0627\u0633 \u0645\u0646 \u062E\u0644\u0627\u0644 \u062D\u0633\u0627\u0628 \u0639\u062F\u062F \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0645\u062A\u0639\u0644\u0642\u0629 \u0628\u0627\u0644\u0628\u062D\u062B \u0639\u0644\u0649 \u0639\u062F\u062F \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0645\u0633\u062A\u0631\u062C\u0639\u0629 \u0627\u0644\u0643\u0644\u064A\u0629. \u0628\u064A\u0646\u0645\u0627 \u0627\u0644\u0625\u0631\u062C\u0627\u0639 \u064A\u0642\u0627\u0633 \u0645\u0646 \u062E\u0644\u0627\u0644 \u062D\u0633\u0627\u0628 \u0639\u062F\u062F \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0645\u062A\u0639\u0644\u0642\u0629 \u0628\u0627\u0644\u0628\u062D\u062B \u0639\u0644\u0649 \u0639\u062F\u062F \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0643\u0644\u064A\u0629.\u0644\u0646\u0641\u0631\u0636 \u0627\u0646 \u0647\u0646\u0627\u0643 \u062C\u0647\u0627\u0632 \u064A\u0633\u062A\u0637\u064A\u0639 \u0627\u0643\u062A\u0634\u0627\u0641 \u0645\u0631\u0636 \u0633\u0631\u0637\u0627\u0646 \u0627\u0644\u0631\u0626\u0629 \u0645\u0646 \u062E\u0644\u0627\u0644 \u0627\u0644\u0635\u0648\u0631. \u0642\u0627\u0645 \u0647\u0630\u0627 \u0627\u0644\u062C\u0647\u0627\u0632 \u0628\u0625\u0631\u062C\u0627\u0639 7 \u0646\u062A\u0627\u0626\u062C, 4 \u0645\u0646\u0647\u0627 \u0646\u062A\u0627\u0626\u062C \u0635\u062D\u064A\u062D\u0629 \u0648\u062B\u0644\u0627\u062B\u0629 \u062E\u0627\u0638\u0626\u0629\u060C \u0648\u0639\u062F\u062F \u0627\u0644\u0635\u0648\u0631 \u0627\u0644\u0643\u0644\u064A\u0629 \u0627\u0644\u062A\u064A \u062A\u062D\u062A\u0648\u064A \u0639\u0644\u0649 \u0645\u0631\u0636 \u0633\u0631\u0637\u0627\u0646 \u0627\u0644\u0631\u0626\u0629 \u0647\u064A 9. \u0641\u064A \u0647\u0630\u0647 \u0627\u0644\u062D\u0627\u0644\u0629 \u0646\u0642\u0648\u0644 \u0623\u0646 \u0645\u0639\u064A\u0627\u0631 \u0627\u0644\u062F\u0642\u0629 \u0647\u0648 4/7 \u0628\u064A\u0646\u0645\u0627 \u0645\u0642\u064A\u0627\u0633 \u0627\u0644\u0625\u0631\u062C\u0627\u0639 \u0647\u0648 4/9.\u0644\u0646\u0623\u062E\u0630 \u0645\u062B\u0627\u0644 \u0622\u062E\u0631: \u0641\u064A \u062D\u0627\u0644\u0629 \u0639\u0645\u0644 \u0639\u0645\u0644\u064A\u0629 \u0628\u062D\u062B \u0639\u0646 \u0633\u064A\u0627\u0631\u0629 \u062C\u0627\u063A\u0648\u0627\u0631\u060C \u0644\u0648 \u0643\u0627\u0646\u062A \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0645\u0631\u062A\u062C\u0639\u0629 \u0647\u064A 20 \u0635\u0648\u0631\u0629 \u0644\u0633\u064A\u0627\u0631\u0629 \u0627\u0644\u062C\u0627\u0639\u0648\u0627\u0631 \u0648\u0639\u0634\u0631\u0629 \u0635\u0648\u0631 \u0644\u062D\u064A\u0648\u0627\u0646 \u0627\u0644\u062C\u0627\u063A\u0648\u0627\u0631 \u062A\u0643\u0648\u0646 \u0627\u0644\u062F\u0642\u0629 20/30 \u0628\u064A\u0646\u0645\u0627 \u0647\u0646\u0627\u0643 40 \u0635\u0648\u0631\u0629 \u0623\u062E\u0631\u0649 \u0644\u0633\u064A\u0627\u0631\u0629 \u0627\u0644\u062C\u0627\u063A\u0648\u0627\u0631 \u0644\u0645 \u064A\u062A\u0645 \u0627\u0633\u062A\u0631\u062C\u0627\u0639\u0647\u0627 \u0645\u0646 \u062E\u0644\u0627\u0644 \u0645\u062D\u0631\u0643 \u0627\u0644\u0628\u062D\u062B\u060C \u0639\u0646\u062F\u0626\u0630 \u0646\u0642\u0648\u0644 \u0623\u0646 \u0645\u0642\u064A\u0627\u0633 \u0627\u0644\u0625\u0631\u062C\u0627\u0639 \u0647\u0648 20/60"@ar . "La precisi\u00F3n y exhaustividad (denominado a veces como exhaustividad y precisi\u00F3n) es una m\u00E9trica empleada en la medida del rendimiento de los sistemas de b\u00FAsqueda y recuperaci\u00F3n de informaci\u00F3n y reconocimiento de patrones. En este contexto se denomina precisi\u00F3n (denominado igualmente valor positivo predicho) como a la fracci\u00F3n de instancias recuperadas que son relevantes, mientras recall (denominado igualmente sensibilidad o exhaustividad) es la fracci\u00F3n de instancias relevantes que han sido recuperadas.\u200B Tanto la precisi\u00F3n como la exhaustividad son entendidas como medidas de la relevancia. Para entender mejor el concepto, supongamos de la existencia de un programa que reconoce perros en fotograf\u00EDas, dicho programa reconoce 7 perros en una escena que contiene 9 perros y algunos gatos. Si 4 "@es . . . . . "Dans les domaines de la reconnaissance de formes, de la recherche d'information et de la classification automatique, la pr\u00E9cision (ou valeur pr\u00E9dictive positive) est la proportion des items pertinents parmi l'ensemble des items propos\u00E9s ; le rappel (ou sensibilit\u00E9) est la proportion des items pertinents propos\u00E9s parmi l'ensemble des items pertinents. Ces deux notions correspondent ainsi \u00E0 une conception et \u00E0 une mesure de la pertinence. Lorsqu'un moteur de recherche, par exemple, retourne 30 pages web dont seulement 20 sont pertinentes (les vrais positifs) et 10 ne le sont pas (les faux positifs), mais qu'il omet 40 autres pages pertinentes (les faux n\u00E9gatifs), sa pr\u00E9cision est de 20/(20+10) = 2/3 et son rappel vaut 20/(20+40) = 1/3. La pr\u00E9cision peut ainsi \u00EAtre comprise comme une mesure de l'exactitude ou de la qualit\u00E9, tandis que le rappel est une mesure de l'exhaustivit\u00E9 ou de la quantit\u00E9."@fr . . . . . . . "In pattern recognition, information retrieval, object detection and classification (machine learning), precision and recall are performance metrics that apply to data retrieved from a collection, corpus or sample space. Precision (also called positive predictive value) is the fraction of relevant instances among the retrieved instances, while recall (also known as sensitivity) is the fraction of relevant instances that were retrieved. Both precision and recall are therefore based on relevance."@en . . "Precision und Recall"@de . . . . . "14343887"^^ . . ", esperimentazioan, diagnostikoan eta informazioaren berreskurapenean doitasuna eta estaldura elementu multzo bat hautematean izandako eraginkortasunaren neurriak dira. Doitasuna hautemandako elementu guztietan zuzen hautemandako elementuen proportzioa da eta estaldura hauteman beharreko elementu guztietatik hautemandako elementuen proportzioa da. Adibidez, espezie bateko 20 aleetatik gaixo daudenak hauteman behar dira eta 8 ale dira gaixotzat jo direnak, horietatik 6 benetako gaixoak izanik; datu horietatik doitasuna 6/8=%75 da (hau da, hautemandako gaixo guztietatik %75 dira benetako gaixoak) eta estaldura 6/20=%30 da (hau da, gaixo guztietatik %30a hauteman da)."@eu . "Doitasun eta estaldura"@eu . "Pr\u00E9cision et rappel"@fr . "23172"^^ . . "Dans les domaines de la reconnaissance de formes, de la recherche d'information et de la classification automatique, la pr\u00E9cision (ou valeur pr\u00E9dictive positive) est la proportion des items pertinents parmi l'ensemble des items propos\u00E9s ; le rappel (ou sensibilit\u00E9) est la proportion des items pertinents propos\u00E9s parmi l'ensemble des items pertinents. Ces deux notions correspondent ainsi \u00E0 une conception et \u00E0 une mesure de la pertinence."@fr . "Precision and recall"@en . . . . . . . . . . "Precisi\u00F3 i reclam"@ca . . "La precisi\u00F3n y exhaustividad (denominado a veces como exhaustividad y precisi\u00F3n) es una m\u00E9trica empleada en la medida del rendimiento de los sistemas de b\u00FAsqueda y recuperaci\u00F3n de informaci\u00F3n y reconocimiento de patrones. En este contexto se denomina precisi\u00F3n (denominado igualmente valor positivo predicho) como a la fracci\u00F3n de instancias recuperadas que son relevantes, mientras recall (denominado igualmente sensibilidad o exhaustividad) es la fracci\u00F3n de instancias relevantes que han sido recuperadas.\u200B Tanto la precisi\u00F3n como la exhaustividad son entendidas como medidas de la relevancia. Para entender mejor el concepto, supongamos de la existencia de un programa que reconoce perros en fotograf\u00EDas, dicho programa reconoce 7 perros en una escena que contiene 9 perros y algunos gatos. Si 4 de las identificaciones han sido correctas, pero 3 eran gatos, el programa tendr\u00E1 una precisi\u00F3n de 4/7 mientras que posee una sensibilidad de 4/9. Otro ejemplo en el que participa un motor de b\u00FAsqueda que, ante una consulta dada, retorna 30 p\u00E1ginas de las cuales s\u00F3lo 20 son relevantes dejando 40 p\u00E1ginas relevantes fuera de la b\u00FAsqueda. Este motor tendr\u00E1 entonces una precisi\u00F3n de 20/30 = 2/3 mientras que su sensibilidad es 20/60 = 1/3. Para un usuario la situaci\u00F3n ideal es aquella en la que existe una precisi\u00F3n y exhaustividad alta (es decir muy cercana a 1). A esta situaci\u00F3n se la denomina utilidad te\u00F3rica. Con el objeto de ponderar y ver cual lejano se encuentran ambas medidas del la utilidad te\u00F3rica, suele emplearse los valores de ambas m\u00E9tricas combinadas en una media arm\u00F3nica denominada valor-F."@es . "Precisione e recupero"@it . . . . . . . "Precisi\u00F3n y exhaustividad"@es . . . . . "Em reconhecimento de padr\u00F5es e recupera\u00E7\u00E3o de informa\u00E7\u00F5es com classifica\u00E7\u00E3o bin\u00E1ria, precis\u00E3o (tamb\u00E9m chamada de valor preditivo positivo) \u00E9 a fra\u00E7\u00E3o de inst\u00E2ncias recuperadas que s\u00E3o relevantes, enquanto revoca\u00E7\u00E3o (tamb\u00E9m conhecida como sensibilidade) \u00E9 a fra\u00E7\u00E3o de inst\u00E2ncias relevantes que s\u00E3o recuperadas. Tanto precis\u00E3o quanto revoca\u00E7\u00E3o (ou recall) s\u00E3o, portanto, bases para o estudo e compreens\u00E3o da medida de relev\u00E2ncia. Suponha que um programa de computador para o reconhecimento de c\u00E3es em cenas de um v\u00EDdeo identifica 7 c\u00E3es em uma cena contendo 9 c\u00E3es e alguns gatos. Se 4 das identifica\u00E7\u00F5es est\u00E3o corretas, mas 3 s\u00E3o, na verdade, gatos, a precis\u00E3o do programa \u00E9 4/7 enquanto a sua revoca\u00E7\u00E3o \u00E9 4/9. Quando um motor de pesquisa retorna 30 p\u00E1ginas mas dessas apenas 20 s\u00E3o relevantes enquan"@pt . . "\u0627\u0644\u062F\u0642\u0629 \u0648\u0627\u0644\u0625\u0631\u062C\u0627\u0639 \u0647\u0648 \u0645\u0641\u0647\u0648\u0645 \u0641\u064A \u0627\u0644\u0631\u064A\u0627\u0636\u064A\u0627\u062A \u0648\u0641\u064A \u0639\u0644\u0645 \u0627\u0644\u062D\u0627\u0633\u0648\u0628 \u0627\u0644\u0645\u062E\u062A\u0635 \u0641\u064A \u0645\u062C\u0627\u0644 \u0627\u0633\u062A\u0631\u062C\u0627\u0639 \u0627\u0644\u0645\u0639\u0644\u0648\u0645\u0627\u062A. \u0627\u0644\u062F\u0642\u0629 \u0647\u064A \u0645\u0639\u064A\u0627\u0631 \u064A\u0642\u0627\u0633 \u0645\u0646 \u062E\u0644\u0627\u0644 \u062D\u0633\u0627\u0628 \u0639\u062F\u062F \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0645\u062A\u0639\u0644\u0642\u0629 \u0628\u0627\u0644\u0628\u062D\u062B \u0639\u0644\u0649 \u0639\u062F\u062F \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0645\u0633\u062A\u0631\u062C\u0639\u0629 \u0627\u0644\u0643\u0644\u064A\u0629. \u0628\u064A\u0646\u0645\u0627 \u0627\u0644\u0625\u0631\u062C\u0627\u0639 \u064A\u0642\u0627\u0633 \u0645\u0646 \u062E\u0644\u0627\u0644 \u062D\u0633\u0627\u0628 \u0639\u062F\u062F \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0645\u062A\u0639\u0644\u0642\u0629 \u0628\u0627\u0644\u0628\u062D\u062B \u0639\u0644\u0649 \u0639\u062F\u062F \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0643\u0644\u064A\u0629.\u0644\u0646\u0641\u0631\u0636 \u0627\u0646 \u0647\u0646\u0627\u0643 \u062C\u0647\u0627\u0632 \u064A\u0633\u062A\u0637\u064A\u0639 \u0627\u0643\u062A\u0634\u0627\u0641 \u0645\u0631\u0636 \u0633\u0631\u0637\u0627\u0646 \u0627\u0644\u0631\u0626\u0629 \u0645\u0646 \u062E\u0644\u0627\u0644 \u0627\u0644\u0635\u0648\u0631. \u0642\u0627\u0645 \u0647\u0630\u0627 \u0627\u0644\u062C\u0647\u0627\u0632 \u0628\u0625\u0631\u062C\u0627\u0639 7 \u0646\u062A\u0627\u0626\u062C, 4 \u0645\u0646\u0647\u0627 \u0646\u062A\u0627\u0626\u062C \u0635\u062D\u064A\u062D\u0629 \u0648\u062B\u0644\u0627\u062B\u0629 \u062E\u0627\u0638\u0626\u0629\u060C \u0648\u0639\u062F\u062F \u0627\u0644\u0635\u0648\u0631 \u0627\u0644\u0643\u0644\u064A\u0629 \u0627\u0644\u062A\u064A \u062A\u062D\u062A\u0648\u064A \u0639\u0644\u0649 \u0645\u0631\u0636 \u0633\u0631\u0637\u0627\u0646 \u0627\u0644\u0631\u0626\u0629 \u0647\u064A 9. \u0641\u064A \u0647\u0630\u0647 \u0627\u0644\u062D\u0627\u0644\u0629 \u0646\u0642\u0648\u0644 \u0623\u0646 \u0645\u0639\u064A\u0627\u0631 \u0627\u0644\u062F\u0642\u0629 \u0647\u0648 4/7 \u0628\u064A\u0646\u0645\u0627 \u0645\u0642\u064A\u0627\u0633 \u0627\u0644\u0625\u0631\u062C\u0627\u0639 \u0647\u0648 4/9.\u0644\u0646\u0623\u062E\u0630 \u0645\u062B\u0627\u0644 \u0622\u062E\u0631: \u0641\u064A \u062D\u0627\u0644\u0629 \u0639\u0645\u0644 \u0639\u0645\u0644\u064A\u0629 \u0628\u062D\u062B \u0639\u0646 \u0633\u064A\u0627\u0631\u0629 \u062C\u0627\u063A\u0648\u0627\u0631\u060C \u0644\u0648 \u0643\u0627\u0646\u062A \u0627\u0644\u0646\u062A\u0627\u0626\u062C \u0627\u0644\u0645\u0631\u062A\u062C\u0639\u0629 \u0647\u064A 20 \u0635\u0648\u0631\u0629 \u0644\u0633\u064A\u0627\u0631\u0629 \u0627\u0644\u062C\u0627\u0639\u0648\u0627\u0631 \u0648\u0639\u0634\u0631\u0629 \u0635\u0648\u0631 \u0644\u062D\u064A\u0648\u0627\u0646 \u0627\u0644\u062C\u0627\u063A\u0648\u0627\u0631 \u062A\u0643\u0648\u0646 \u0627\u0644\u062F\u0642\u0629 20/30 \u0628\u064A\u0646\u0645\u0627 \u0647\u0646\u0627\u0643 40 \u0635\u0648\u0631\u0629 \u0623\u062E\u0631\u0649 \u0644\u0633\u064A\u0627\u0631\u0629 \u0627\u0644\u062C\u0627\u063A\u0648\u0627\u0631 \u0644\u0645 \u064A\u062A\u0645 \u0627\u0633\u062A\u0631\u062C\u0627\u0639\u0647\u0627 \u0645\u0646 \u062E\u0644\u0627\u0644 \u0645\u062D\u0631\u0643 \u0627\u0644\u0628\u062D\u062B\u060C \u0639\u0646\u062F\u0626\u0630 \u0646\u0642\u0648\u0644 \u0623\u0646 \u0645\u0642\u064A\u0627\u0633 \u0627\u0644\u0625\u0631\u062C\u0627\u0639 \u0647\u0648 20/60"@ar . . . "Precisione e recupero, o richiamo (in inglese precision e recall) sono due comuni classificazioni statistiche, utilizzate in diversi ambiti del sapere, come per es. l'information retrieval. La precisione pu\u00F2 essere vista come una misura di esattezza o fedelt\u00E0, mentre il recupero \u00E8 una misura di completezza. Nell'Information Retrieval, la precisione \u00E8 definita come il numero di documenti attinenti recuperati da una ricerca diviso il numero totale di documenti recuperati dalla stessa ricerca, e il recupero \u00E8 definito come il numero di documenti attinenti recuperati da una ricerca diviso il numero totale di documenti attinenti esistenti (che dovrebbe essere stato recuperato). In un processo di classificazione statistica, la precisione per una classe \u00E8 il numero di veri positivi (il numero di oggetti etichettati correttamente come appartenenti alla classe) diviso il numero totale di elementi etichettati come appartenenti alla classe (la somma di veri positivi e falsi positivi, che sono oggetti etichettati erroneamente come appartenenti alla classe).Recupero in questo contesto \u00E8 definito come il numero di veri positivi diviso il numero totale di elementi che effettivamente appartengono alla classe (per esempio la somma di veri positivi e falsi negativi, che sono oggetti che non sono stati etichettati come appartenenti alla classe ma dovrebbero esserlo). Nell'Information Retrieval, un valore di precisione di 1.0 significa che ogni risultato recuperato da una ricerca \u00E8 attinente mentre un valore di recupero pari a 1.0 significa che tutti i documenti attinenti sono stati recuperati dalla ricerca. In un processo di classificazione, un valore di precisione di 1.0 per la classe C significa che ogni oggetto che \u00E8 stato etichettato come appartenente alla classe C vi appartiene davvero (ma non dice niente sul numero di elementi della classe C che non sono stati etichettati correttamente) mentre un valore di recupero pari ad 1.0 significa che ogni oggetto della classe C \u00E8 stato etichettato come appartenente ad essa (ma non dice niente sul numero di elementi etichettati non correttamente con C)."@it . . "1122267443"^^ . . "Em reconhecimento de padr\u00F5es e recupera\u00E7\u00E3o de informa\u00E7\u00F5es com classifica\u00E7\u00E3o bin\u00E1ria, precis\u00E3o (tamb\u00E9m chamada de valor preditivo positivo) \u00E9 a fra\u00E7\u00E3o de inst\u00E2ncias recuperadas que s\u00E3o relevantes, enquanto revoca\u00E7\u00E3o (tamb\u00E9m conhecida como sensibilidade) \u00E9 a fra\u00E7\u00E3o de inst\u00E2ncias relevantes que s\u00E3o recuperadas. Tanto precis\u00E3o quanto revoca\u00E7\u00E3o (ou recall) s\u00E3o, portanto, bases para o estudo e compreens\u00E3o da medida de relev\u00E2ncia. Suponha que um programa de computador para o reconhecimento de c\u00E3es em cenas de um v\u00EDdeo identifica 7 c\u00E3es em uma cena contendo 9 c\u00E3es e alguns gatos. Se 4 das identifica\u00E7\u00F5es est\u00E3o corretas, mas 3 s\u00E3o, na verdade, gatos, a precis\u00E3o do programa \u00E9 4/7 enquanto a sua revoca\u00E7\u00E3o \u00E9 4/9. Quando um motor de pesquisa retorna 30 p\u00E1ginas mas dessas apenas 20 s\u00E3o relevantes enquanto deixa de retornar 40 outras p\u00E1ginas relevantes a precis\u00E3o \u00E9 de 20/30 = 2/3 enquanto a revocabilidade 20/60 = 1/3. Precis\u00E3o, neste caso, \u00E9 ''o quanto os resultados da pesquisa s\u00E3o \u00FAteis'', enquanto revocabilidade \u00E9 ''o qu\u00E3o completos os resultados est\u00E3o''. Em estat\u00EDstica, se a hip\u00F3tese nula \u00E9 de que todos (e somente) os itens relevantes s\u00E3o recuperados, a aus\u00EAncia de erros tipo I e tipo II correspondem, respectivamente, \u00E0 precis\u00E3o m\u00E1xima (sem falsos positivos) e revocabilidade m\u00E1xima (sem falsos negativos). No exemplo apresentado, o padr\u00E3o de reconhecimento contem 7 - 4 = 3 erros tipo I e 9 - 4 = 5 erros tipo II. Precis\u00E3o pode ser traduzida como uma medida de exatid\u00E3o ou qualidade, enquanto que revocabilidade \u00E9 a medida de completude ou quantidade. Em termos simples, alta precis\u00E3o significa que o algoritmo retornou substancialmente mais resultados relevantes que irrelevantes, enquanto alta revocabilidade significa que o algoritmo retornou a maioria dos resultados relevantes."@pt . . . "Precis\u00E3o e revoca\u00E7\u00E3o"@pt . "\u0412 \u0440\u043E\u0437\u043F\u0456\u0437\u043D\u0430\u0432\u0430\u043D\u043D\u0456 \u043E\u0431\u0440\u0430\u0437\u0456\u0432, \u0456\u043D\u0444\u043E\u0440\u043C\u0430\u0446\u0456\u0439\u043D\u043E\u043C\u0443 \u043F\u043E\u0448\u0443\u043A\u0443 \u0442\u0430 \u043A\u043B\u0430\u0441\u0438\u0444\u0456\u043A\u0430\u0446\u0456\u0457, \u0432\u043B\u0443\u0301\u0447\u043D\u0456\u0441\u0442\u044C (\u0430\u043D\u0433\u043B. precision, \u044F\u043A\u0443 \u0442\u0430\u043A\u043E\u0436 \u043D\u0430\u0437\u0438\u0432\u0430\u044E\u0442\u044C \u043F\u0440\u043E\u0433\u043D\u043E\u0441\u0442\u0438\u0447\u043D\u043E\u044E \u0437\u043D\u0430\u0447\u0443\u0449\u0456\u0441\u0442\u044E \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u043E\u0433\u043E \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0443) \u0454 \u0447\u0430\u0441\u0442\u043A\u043E\u044E \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u0445 \u0437\u0440\u0430\u0437\u043A\u0456\u0432 \u0441\u0435\u0440\u0435\u0434 \u0437\u043D\u0430\u0439\u0434\u0435\u043D\u0438\u0445, \u0442\u043E\u0434\u0456 \u044F\u043A \u043F\u043E\u0432\u043D\u043E\u0442\u0430\u0301 (\u0430\u043D\u0433\u043B. recall, \u0432\u0456\u0434\u043E\u043C\u0430 \u0442\u0430\u043A\u043E\u0436 \u044F\u043A \u0447\u0443\u0442\u043B\u0438\u0432\u0456\u0441\u0442\u044C) \u0454 \u0447\u0430\u0441\u0442\u043A\u043E\u044E \u0437\u0430\u0433\u0430\u043B\u044C\u043D\u043E\u0433\u043E \u0447\u0438\u0441\u043B\u0430 \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0438\u0445 \u0437\u0440\u0430\u0437\u043A\u0456\u0432, \u044F\u043A\u0443 \u0431\u0443\u043B\u043E \u0434\u0456\u0439\u0441\u043D\u043E \u0437\u043D\u0430\u0439\u0434\u0435\u043D\u043E. \u042F\u043A \u0432\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C, \u0442\u0430\u043A \u0456 \u043F\u043E\u0432\u043D\u043E\u0442\u0430, \u0432\u0456\u0434\u0442\u0430\u043A \u0491\u0440\u0443\u043D\u0442\u0443\u044E\u0442\u044C\u0441\u044F \u043D\u0430 \u0440\u043E\u0437\u0443\u043C\u0456\u043D\u043D\u0456 \u0442\u0430 \u043C\u0456\u0440\u0456 \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u043E\u0441\u0442\u0456. \u0412\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u043D\u0435 \u0441\u043B\u0456\u0434 \u043F\u043B\u0443\u0442\u0430\u0442\u0438 \u0437 (\u0430\u043D\u0433\u043B. accuracy), \u044F\u043A\u0430 \u0454 \u0447\u0430\u0441\u0442\u043A\u043E\u044E \u043F\u0440\u0430\u0432\u0438\u043B\u044C\u043D\u043E \u0441\u043F\u0440\u043E\u0433\u043D\u043E\u0437\u043E\u0432\u0430\u043D\u0438\u0445 \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0456\u0432, \u044F\u043A \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0438\u0445, \u0442\u0430\u043A \u0456 \u043D\u0435\u0433\u0430\u0442\u0438\u0432\u043D\u0438\u0445. \u0412\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u0441\u0442\u043E\u0441\u0443\u0454\u0442\u044C\u0441\u044F \u043B\u0438\u0448\u0435 \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0438\u0445 \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0456\u0432. \u041D\u0435\u0445\u0430\u0439 \u043A\u043E\u043C\u043F'\u044E\u0442\u0435\u0440\u043D\u0430 \u043F\u0440\u043E\u0433\u0440\u0430\u043C\u0430 \u0434\u043B\u044F \u0440\u043E\u0437\u043F\u0456\u0437\u043D\u0430\u0432\u0430\u043D\u043D\u044F \u0441\u043E\u0431\u0430\u043A \u043D\u0430 \u0444\u043E\u0442\u043E\u0433\u0440\u0430\u0444\u0456\u044F\u0445 \u0432\u0438\u044F\u0432\u0438\u043B\u0430 8 \u0441\u043E\u0431\u0430\u043A \u043D\u0430 \u0437\u043E\u0431\u0440\u0430\u0436\u0435\u043D\u043D\u0456, \u0449\u043E \u043C\u0456\u0441\u0442\u0438\u0442\u044C 10 \u043A\u043E\u0442\u0456\u0432 \u0442\u0430 12 \u0441\u043E\u0431\u0430\u043A (\u0432\u043B\u0430\u0441\u043D\u0435 \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u0445 \u0435\u043B\u0435\u043C\u0435\u043D\u0442\u0456\u0432). \u0421\u0435\u0440\u0435\u0434 8 \u0456\u0434\u0435\u043D\u0442\u0438\u0444\u0456\u043A\u043E\u0432\u0430\u043D\u0438\u0445 \u044F\u043A \u0441\u043E\u0431\u0430\u043A\u0438 5 \u0456 \u0441\u043F\u0440\u0430\u0432\u0434\u0456 \u0454 \u0441\u043E\u0431\u0430\u043A\u0430\u043C\u0438 (\u0456\u0441\u0442\u0438\u043D\u043D\u043E \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0456), \u0442\u043E\u0434\u0456 \u044F\u043A \u0456\u043D\u0448\u0456 3 \u0454 \u043A\u043E\u0442\u0430\u043C\u0438 (\u0445\u0438\u0431\u043D\u043E \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0456). 7 \u0441\u043E\u0431\u0430\u043A \u0431\u0443\u043B\u043E \u043F\u0440\u043E\u043F\u0443\u0449\u0435\u043D\u043E (\u0445\u0438\u0431\u043D\u043E \u043D\u0435\u0433\u0430\u0442\u0438\u0432\u043D\u0456), \u0430 7 \u043A\u043E\u0442\u0456\u0432 \u0431\u0443\u043B\u043E \u0432\u0438\u043A\u043B\u044E\u0447\u0435\u043D\u043E \u043F\u0440\u0430\u0432\u0438\u043B\u044C\u043D\u043E (\u0456\u0441\u0442\u0438\u043D\u043D\u043E \u043D\u0435\u0433\u0430\u0442\u0438\u0432\u043D\u0456). \u0412\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u0446\u0456\u0454\u0457 \u043F\u0440\u043E\u0433\u0440\u0430\u043C\u0438 \u0441\u0442\u0430\u043D\u043E\u0432\u0438\u0442\u044C 5/8 (\u0456\u0441\u0442\u0438\u043D\u043D\u043E \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0456 / \u0432\u0441\u0456 \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0456), \u0442\u043E\u0434\u0456 \u044F\u043A \u043F\u043E\u0432\u043D\u043E\u0442\u0430 \u2014 5/12 (\u0456\u0441\u0442\u0438\u043D\u043D\u043E \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0456 / \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0456 \u0435\u043B\u0435\u043C\u0435\u043D\u0442\u0438). \u041A\u043E\u043B\u0438 \u043F\u043E\u0448\u0443\u043A\u043E\u0432\u0438\u0439 \u0440\u0443\u0448\u0456\u0439 \u043F\u043E\u0432\u0435\u0440\u0442\u0430\u0454 30 \u0441\u0442\u043E\u0440\u0456\u043D\u043E\u043A, \u043B\u0438\u0448\u0435 20 \u0437 \u044F\u043A\u0438\u0445 \u0454 \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u043C\u0438, \u0432 \u0442\u043E\u0439 \u0436\u0435 \u0447\u0430\u0441 \u0432\u0438\u044F\u0432\u043B\u044F\u044E\u0447\u0438\u0441\u044C \u043D\u0435\u0437\u0434\u0430\u0442\u043D\u0438\u043C \u043F\u043E\u0432\u0435\u0440\u043D\u0443\u0442\u0438 40 \u0434\u043E\u0434\u0430\u0442\u043A\u043E\u0432\u0438\u0445 \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u0445 \u0441\u0442\u043E\u0440\u0456\u043D\u043E\u043A, \u0439\u043E\u0433\u043E \u0432\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u0441\u0442\u0430\u043D\u043E\u0432\u0438\u0442\u044C 20/30 = 2/3, \u0442\u043E\u0434\u0456 \u044F\u043A \u043F\u043E\u0432\u043D\u043E\u0442\u0430 \u2014 20/60 = 1/3. \u0422\u043E\u0436, \u0443 \u0446\u044C\u043E\u043C\u0443 \u0432\u0438\u043F\u0430\u0434\u043A\u0443, \u0432\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u043F\u043E\u043A\u0430\u0437\u0443\u0454, \u00AB\u043D\u0430\u0441\u043A\u0456\u043B\u044C\u043A\u0438 \u043F\u0440\u0430\u0432\u0438\u043B\u044C\u043D\u0438\u043C\u0438 \u0454 \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0438\u00BB, \u0442\u043E\u0434\u0456 \u044F\u043A \u043F\u043E\u0432\u043D\u043E\u0442\u0430 \u2014 \u00AB\u043D\u0430\u0441\u043A\u0456\u043B\u044C\u043A\u0438 \u043F\u043E\u0432\u043D\u0438\u043C\u0438 \u0454 \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0438\u00BB. \u042F\u043A\u0449\u043E \u0437\u0430\u0441\u0442\u043E\u0441\u043E\u0432\u0443\u0432\u0430\u0442\u0438 \u043F\u0456\u0434\u0445\u0456\u0434 \u043F\u0435\u0440\u0435\u0432\u0456\u0440\u043A\u0438 \u0433\u0456\u043F\u043E\u0442\u0435\u0437 \u0437\u0456 \u0441\u0442\u0430\u0442\u0438\u0441\u0442\u0438\u043A\u0438, \u0432 \u044F\u043A\u043E\u043C\u0443, \u0432 \u0446\u044C\u043E\u043C\u0443 \u0432\u0438\u043F\u0430\u0434\u043A\u0443, \u043D\u0443\u043B\u044C\u043E\u0432\u0430 \u0433\u0456\u043F\u043E\u0442\u0435\u0437\u0430 \u043F\u043E\u043B\u044F\u0433\u0430\u0454 \u0432 \u0442\u0456\u043C, \u0449\u043E \u0437\u0430\u0434\u0430\u043D\u0438\u0439 \u0437\u0440\u0430\u0437\u043E\u043A \u0454 \u043D\u0435\u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u043C, \u0442\u043E\u0431\u0442\u043E, \u043D\u0435 \u0441\u043E\u0431\u0430\u043A\u043E\u044E, \u0442\u043E \u0432\u0456\u0434\u0441\u0443\u0442\u043D\u0456\u0441\u0442\u044C \u043F\u043E\u043C\u0438\u043B\u043E\u043A \u043F\u0435\u0440\u0448\u043E\u0433\u043E \u0456 \u0434\u0440\u0443\u0433\u043E\u0433\u043E \u0440\u043E\u0434\u0443 (\u0442\u043E\u0431\u0442\u043E, \u0456\u0434\u0435\u0430\u043B\u044C\u043D\u0456 \u0447\u0443\u0442\u043B\u0438\u0432\u0456\u0441\u0442\u044C \u0442\u0430 \u0441\u043F\u0435\u0446\u0438\u0444\u0456\u0447\u043D\u0456\u0441\u0442\u044C \u0443 100 % \u043A\u043E\u0436\u043D\u0430) \u0432\u0456\u0434\u043F\u043E\u0432\u0456\u0434\u0430\u0454, \u0432\u0456\u0434\u043F\u043E\u0432\u0456\u0434\u043D\u043E, \u0456\u0434\u0435\u0430\u043B\u044C\u043D\u0456\u0439 \u0432\u043B\u0443\u0447\u043D\u043E\u0441\u0442\u0456 (\u0431\u0435\u0437 \u0445\u0438\u0431\u043D\u043E \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0438\u0445) \u0442\u0430 \u0456\u0434\u0435\u0430\u043B\u044C\u043D\u0456\u0439 \u043F\u043E\u0432\u043D\u043E\u0442\u0456 (\u0431\u0435\u0437 \u0445\u0438\u0431\u043D\u043E \u043D\u0435\u0433\u0430\u0442\u0438\u0432\u043D\u0438\u0445). \u0417\u0430\u0433\u0430\u043B\u044C\u043D\u0456\u0448\u0435, \u043F\u043E\u0432\u043D\u043E\u0442\u0430 \u0454 \u043F\u0440\u043E\u0441\u0442\u043E \u0434\u043E\u043F\u043E\u0432\u043D\u0435\u043D\u043D\u044F\u043C \u0440\u0456\u0432\u043D\u044F \u043F\u043E\u043C\u0438\u043B\u043E\u043A II \u0440\u043E\u0434\u0443, \u0442\u043E\u0431\u0442\u043E, \u043E\u0434\u0438\u043D\u0438\u0446\u044F \u043C\u0456\u043D\u0443\u0441 \u0440\u0456\u0432\u0435\u043D\u044C \u043F\u043E\u043C\u0438\u043B\u043E\u043A II \u0440\u043E\u0434\u0443. \u0412\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u043F\u043E\u0432'\u044F\u0437\u0430\u043D\u0430 \u0437 \u0440\u0456\u0432\u043D\u0435\u043C \u043F\u043E\u043C\u0438\u043B\u043E\u043A I \u0440\u043E\u0434\u0443, \u0430\u043B\u0435 \u0434\u0435\u0449\u043E \u0441\u043A\u043B\u0430\u0434\u043D\u0456\u0448\u0438\u043C \u0447\u0438\u043D\u043E\u043C, \u043E\u0441\u043A\u0456\u043B\u044C\u043A\u0438 \u0432\u043E\u043D\u0430 \u0442\u0430\u043A\u043E\u0436 \u0437\u0430\u043B\u0435\u0436\u0438\u0442\u044C \u0432\u0456\u0434 \u0430\u043F\u0440\u0456\u043E\u0440\u043D\u043E\u0433\u043E \u0440\u043E\u0437\u043F\u043E\u0434\u0456\u043B\u0443 \u0441\u043F\u043E\u0441\u0442\u0435\u0440\u0456\u0433\u0430\u043D\u043D\u044F \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u043E\u0433\u043E, \u0447\u0438 \u043D\u0435\u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u043E\u0433\u043E \u0437\u0440\u0430\u0437\u043A\u0430. \u041D\u0430\u0432\u0435\u0434\u0435\u043D\u0438\u0439 \u0432\u0438\u0449\u0435 \u043F\u0440\u0438\u043A\u043B\u0430\u0434 \u0456\u0437 \u043A\u043E\u0442\u0430\u043C\u0438 \u0442\u0430 \u0441\u043E\u0431\u0430\u043A\u0430\u043C\u0438 \u043C\u0456\u0441\u0442\u0438\u0432 8 \u2212 5 = 3 \u043F\u043E\u043C\u0438\u043B\u043A\u0438 I \u0440\u043E\u0434\u0443, \u0449\u043E \u0434\u0430\u0432\u0430\u043B\u043E \u0440\u0456\u0432\u0435\u043D\u044C \u043F\u043E\u043C\u0438\u043B\u043E\u043A I \u0440\u043E\u0434\u0443 3/10, \u0442\u0430 12 \u2212 5 = 7 \u043F\u043E\u043C\u0438\u043B\u043E\u043A II \u0440\u043E\u0434\u0443, \u0449\u043E \u0434\u0430\u0432\u0430\u043B\u043E \u0440\u0456\u0432\u0435\u043D\u044C \u043F\u043E\u043C\u0438\u043B\u043E\u043A II \u0440\u043E\u0434\u0443 7/12. \u0412\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u043C\u043E\u0436\u043B\u0438\u0432\u043E \u0440\u043E\u0437\u0433\u043B\u044F\u0434\u0430\u0442\u0438 \u044F\u043A \u0440\u0456\u0432\u0435\u043D\u044C \u044F\u043A\u043E\u0441\u0442\u0456, \u0442\u043E\u0434\u0456 \u044F\u043A \u043F\u043E\u0432\u043D\u043E\u0442\u0443 \u2014 \u044F\u043A \u0440\u0456\u0432\u0435\u043D\u044C \u043A\u0456\u043B\u044C\u043A\u043E\u0441\u0442\u0456. \u0412\u0438\u0449\u0430 \u0432\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u043E\u0437\u043D\u0430\u0447\u0430\u0454, \u0449\u043E \u0430\u043B\u0433\u043E\u0440\u0438\u0442\u043C \u0432\u0438\u0434\u0430\u0454 \u0431\u0456\u043B\u044C\u0448\u0435 \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u0445 \u0437\u0440\u0430\u0437\u043A\u0456\u0432, \u043D\u0456\u0436 \u043D\u0435\u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u0445, \u0430 \u0432\u0438\u0441\u043E\u043A\u0430 \u043F\u043E\u0432\u043D\u043E\u0442\u0430 \u043E\u0437\u043D\u0430\u0447\u0430\u0454, \u0449\u043E \u0430\u043B\u0433\u043E\u0440\u0438\u0442\u043C \u0432\u0438\u0434\u0430\u0454 \u0431\u0456\u043B\u044C\u0448\u0456\u0441\u0442\u044C \u0456\u0437 \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u0445 \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0456\u0432 (\u043D\u0435\u0437\u0430\u043B\u0435\u0436\u043D\u043E \u0432\u0456\u0434 \u0442\u043E\u0433\u043E, \u0447\u0438 \u0432\u0456\u043D \u0442\u0430\u043A\u043E\u0436 \u0432\u0438\u0434\u0430\u0454 \u0439 \u043D\u0435\u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0456)."@uk . . . . . . . . . . . . . . . . . "\u0412 \u0440\u043E\u0437\u043F\u0456\u0437\u043D\u0430\u0432\u0430\u043D\u043D\u0456 \u043E\u0431\u0440\u0430\u0437\u0456\u0432, \u0456\u043D\u0444\u043E\u0440\u043C\u0430\u0446\u0456\u0439\u043D\u043E\u043C\u0443 \u043F\u043E\u0448\u0443\u043A\u0443 \u0442\u0430 \u043A\u043B\u0430\u0441\u0438\u0444\u0456\u043A\u0430\u0446\u0456\u0457, \u0432\u043B\u0443\u0301\u0447\u043D\u0456\u0441\u0442\u044C (\u0430\u043D\u0433\u043B. precision, \u044F\u043A\u0443 \u0442\u0430\u043A\u043E\u0436 \u043D\u0430\u0437\u0438\u0432\u0430\u044E\u0442\u044C \u043F\u0440\u043E\u0433\u043D\u043E\u0441\u0442\u0438\u0447\u043D\u043E\u044E \u0437\u043D\u0430\u0447\u0443\u0449\u0456\u0441\u0442\u044E \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u043E\u0433\u043E \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0443) \u0454 \u0447\u0430\u0441\u0442\u043A\u043E\u044E \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u0438\u0445 \u0437\u0440\u0430\u0437\u043A\u0456\u0432 \u0441\u0435\u0440\u0435\u0434 \u0437\u043D\u0430\u0439\u0434\u0435\u043D\u0438\u0445, \u0442\u043E\u0434\u0456 \u044F\u043A \u043F\u043E\u0432\u043D\u043E\u0442\u0430\u0301 (\u0430\u043D\u0433\u043B. recall, \u0432\u0456\u0434\u043E\u043C\u0430 \u0442\u0430\u043A\u043E\u0436 \u044F\u043A \u0447\u0443\u0442\u043B\u0438\u0432\u0456\u0441\u0442\u044C) \u0454 \u0447\u0430\u0441\u0442\u043A\u043E\u044E \u0437\u0430\u0433\u0430\u043B\u044C\u043D\u043E\u0433\u043E \u0447\u0438\u0441\u043B\u0430 \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0438\u0445 \u0437\u0440\u0430\u0437\u043A\u0456\u0432, \u044F\u043A\u0443 \u0431\u0443\u043B\u043E \u0434\u0456\u0439\u0441\u043D\u043E \u0437\u043D\u0430\u0439\u0434\u0435\u043D\u043E. \u042F\u043A \u0432\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C, \u0442\u0430\u043A \u0456 \u043F\u043E\u0432\u043D\u043E\u0442\u0430, \u0432\u0456\u0434\u0442\u0430\u043A \u0491\u0440\u0443\u043D\u0442\u0443\u044E\u0442\u044C\u0441\u044F \u043D\u0430 \u0440\u043E\u0437\u0443\u043C\u0456\u043D\u043D\u0456 \u0442\u0430 \u043C\u0456\u0440\u0456 \u0440\u0435\u043B\u0435\u0432\u0430\u043D\u0442\u043D\u043E\u0441\u0442\u0456. \u0412\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u043D\u0435 \u0441\u043B\u0456\u0434 \u043F\u043B\u0443\u0442\u0430\u0442\u0438 \u0437 (\u0430\u043D\u0433\u043B. accuracy), \u044F\u043A\u0430 \u0454 \u0447\u0430\u0441\u0442\u043A\u043E\u044E \u043F\u0440\u0430\u0432\u0438\u043B\u044C\u043D\u043E \u0441\u043F\u0440\u043E\u0433\u043D\u043E\u0437\u043E\u0432\u0430\u043D\u0438\u0445 \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0456\u0432, \u044F\u043A \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0438\u0445, \u0442\u0430\u043A \u0456 \u043D\u0435\u0433\u0430\u0442\u0438\u0432\u043D\u0438\u0445. \u0412\u043B\u0443\u0447\u043D\u0456\u0441\u0442\u044C \u0441\u0442\u043E\u0441\u0443\u0454\u0442\u044C\u0441\u044F \u043B\u0438\u0448\u0435 \u043F\u043E\u0437\u0438\u0442\u0438\u0432\u043D\u0438\u0445 \u0440\u0435\u0437\u0443\u043B\u044C\u0442\u0430\u0442\u0456\u0432."@uk . . "En reconeixement de patrons, recuperaci\u00F3 d'informaci\u00F3 i classificaci\u00F3 (aprenentatge autom\u00E0tic), la precisi\u00F3 \u00E9s la fracci\u00F3 de casos veritablement positius entre els casos seleccionats com a positius. El reclam (tamb\u00E9 anomenat sensibilitat, quan es compara amb l'especificitat) \u00E9s la fracci\u00F3 de casos veritablement positius entre els casos rellevants. Tant la precisi\u00F3 com el reclam s\u00F3n mesures de la rellev\u00E0ncia. Suposem que un programa d'ordinador per a recon\u00E8ixer gossos en fotografies identifica 8 gossos a una fotografia que cont\u00E9 12 gossos i alguns gats. Dels 8 identificats com a gossos, 5 s\u00F3n realment gossos (veritables positius), mentre que la resta s\u00F3n gats (falsos positius). La precisi\u00F3 del programa \u00E9s 5/8 i el reclam \u00E9s 5/12. O suposem un motor de cerca que retorna 30 p\u00E0gines. D'aquestes, nom\u00E9s 20 s\u00F3n rellevants, mentre que en falten 40 de rellevants per retornar. La precisi\u00F3 del motor \u00E9s 20/30 = 2/3 i el reclam \u00E9s 20/60 = 1/3. Aix\u00ED doncs, en aquest cas, la precisi\u00F3 \u00E9s \"quant d'\u00FAtils s\u00F3n els resultats\" i el reclam \u00E9s \"quant de complets s\u00F3n els resultats\"."@ca . "Precisione e recupero, o richiamo (in inglese precision e recall) sono due comuni classificazioni statistiche, utilizzate in diversi ambiti del sapere, come per es. l'information retrieval. La precisione pu\u00F2 essere vista come una misura di esattezza o fedelt\u00E0, mentre il recupero \u00E8 una misura di completezza. Nell'Information Retrieval, un valore di precisione di 1.0 significa che ogni risultato recuperato da una ricerca \u00E8 attinente mentre un valore di recupero pari a 1.0 significa che tutti i documenti attinenti sono stati recuperati dalla ricerca."@it . "In pattern recognition, information retrieval, object detection and classification (machine learning), precision and recall are performance metrics that apply to data retrieved from a collection, corpus or sample space. Precision (also called positive predictive value) is the fraction of relevant instances among the retrieved instances, while recall (also known as sensitivity) is the fraction of relevant instances that were retrieved. Both precision and recall are therefore based on relevance. Consider a computer program for recognizing dogs (the relevant element) in a digital photograph. Upon processing a picture which contains ten cats and twelve dogs, the program identifies eight dogs. Of the eight elements identified as dogs, only five actually are dogs (true positives), while the other three are cats (false positives). Seven dogs were missed (false negatives), and seven cats were correctly excluded (true negatives). The program's precision is then 5/8 (true positives / selected elements) while its recall is 5/12 (true positives / relevant elements). When a search engine returns 30 pages, only 20 of which are relevant, while failing to return 40 additional relevant pages, its precision is 20/30 = 2/3, which tells us how valid the results are, while its recall is 20/60 = 1/3, which tells us how complete the results are. Adopting a hypothesis-testing approach from statistics, in which, in this case, the null hypothesis is that a given item is irrelevant, i.e., not a dog, absence of type I and type II errors (i.e. perfect specificity and sensitivity of 100% each) corresponds respectively to perfect precision (no false positive) and perfect recall (no false negative). More generally, recall is simply the complement of the type II error rate, i.e. one minus the type II error rate. Precision is related to the type I error rate, but in a slightly more complicated way, as it also depends upon the prior distribution of seeing a relevant vs an irrelevant item. The above cat and dog example contained 8 \u2212 5 = 3 type I errors (false positives) out of 10 total cats (true negatives), for a type I error rate of 3/10, and 12 \u2212 5 = 7 type II errors, for a type II error rate of 7/12. Precision can be seen as a measure of quality, and recall as a measure of quantity. Higher precision means that an algorithm returns more relevant results than irrelevant ones, and high recall means that an algorithm returns most of the relevant results (whether or not irrelevant ones are also returned)."@en . "\uC774\uC9C4 \uBD84\uB958 \uAE30\uBC95(binary classification)\uC744 \uC0AC\uC6A9\uD558\uB294 \uD328\uD134 \uC778\uC2DD\uACFC \uC815\uBCF4 \uAC80\uC0C9 \uBD84\uC57C\uC5D0\uC11C, \uC815\uBC00\uB3C4\uB294 \uAC80\uC0C9\uB41C \uACB0\uACFC\uB4E4 \uC911 \uAD00\uB828 \uC788\uB294 \uAC83\uC73C\uB85C \uBD84\uB958\uB41C \uACB0\uACFC\uBB3C\uC758 \uBE44\uC728\uC774\uACE0, \uC7AC\uD604\uC728\uC740 \uAD00\uB828 \uC788\uB294 \uAC83\uC73C\uB85C \uBD84\uB958\uB41C \uD56D\uBAA9\uB4E4 \uC911 \uC2E4\uC81C \uAC80\uC0C9\uB41C \uD56D\uBAA9\uB4E4\uC758 \uBE44\uC728\uC774\uB2E4. \uB530\uB77C\uC11C \uC815\uBC00\uB3C4\uC640 \uC7AC\uD604\uC728 \uBAA8\uB450 \uAD00\uB828\uB3C4(Relevance)\uC758 \uCE21\uC815 \uAE30\uC900 \uBC0F \uC9C0\uC2DD\uC744 \uD1A0\uB300\uB85C \uD558\uACE0 \uC788\uB2E4."@ko . "\uC815\uBC00\uB3C4\uC640 \uC7AC\uD604\uC728"@ko . . . . . . "En reconeixement de patrons, recuperaci\u00F3 d'informaci\u00F3 i classificaci\u00F3 (aprenentatge autom\u00E0tic), la precisi\u00F3 \u00E9s la fracci\u00F3 de casos veritablement positius entre els casos seleccionats com a positius. El reclam (tamb\u00E9 anomenat sensibilitat, quan es compara amb l'especificitat) \u00E9s la fracci\u00F3 de casos veritablement positius entre els casos rellevants. Tant la precisi\u00F3 com el reclam s\u00F3n mesures de la rellev\u00E0ncia."@ca . .