By the modification of the 111th article of the Constitution, the reasonable legal trial duration is a civil right for everyone, according to the article 6th of the European Convention on human rights, ratified in Italy by the law n. 848/1995. On march 24th 2001, Italian Parliament has approved the so called Pinto law, n. 89/2001, not to get early definitions, but just to compensate the damage from a not reasonable legal trial duration. This law states one who complains the excessive legal trial duration can appeal in Italy to Appeals Court and get, on charge of Italian government, the compensation for moral and patrimonial damages coming from his excessive legal trial duration. Jurisdictional product is not the same as industrial product. Court cases are defined ‘simple’ or ‘difficult’ and imply a different quality and quantity of judge work depending on the consideration of many variables (number of parts, hearings, etc.). Thanks to the progressive court computerization, the process to extract knowledge from large databases, i.e. Data Mining, becomes possible for judi-cial office managers. At first it is important to define a shared and agreed efficiency and efficacy indicators panel, such to consider all the quantitative and qualitative variables involved in the programming activity. Judicial flows evaluation and monitoring by multidimensional and multivariate statistical analysis becomes necessary to overcome the limits of simple descriptive and univariate statistics. In order to study the multivariate dependence problem of legal trial dura-tion in judicial flows, in this work we focus on a generalization of Partial Least Squares (PLS, Wold, 1966; Tenenhaus, 1998) called Multivariate Additive PLS via Splines (MAPLSS, Lombardo et al., 2007). MAPLSS permits to consider non-linearity and it is particularly opportune to capture interactions between variables of different nature, as the interaction itself, for example between the number of pending proceedings and the number of hearings, can affect the duration of trials more than the simple main effects (pending proceedings or number of hearing) separately do.
Data mining and MAPLSS regression tools. A case study: the evaluation of legal trial duration
LOMBARDO, Rosaria;
2009
Abstract
By the modification of the 111th article of the Constitution, the reasonable legal trial duration is a civil right for everyone, according to the article 6th of the European Convention on human rights, ratified in Italy by the law n. 848/1995. On march 24th 2001, Italian Parliament has approved the so called Pinto law, n. 89/2001, not to get early definitions, but just to compensate the damage from a not reasonable legal trial duration. This law states one who complains the excessive legal trial duration can appeal in Italy to Appeals Court and get, on charge of Italian government, the compensation for moral and patrimonial damages coming from his excessive legal trial duration. Jurisdictional product is not the same as industrial product. Court cases are defined ‘simple’ or ‘difficult’ and imply a different quality and quantity of judge work depending on the consideration of many variables (number of parts, hearings, etc.). Thanks to the progressive court computerization, the process to extract knowledge from large databases, i.e. Data Mining, becomes possible for judi-cial office managers. At first it is important to define a shared and agreed efficiency and efficacy indicators panel, such to consider all the quantitative and qualitative variables involved in the programming activity. Judicial flows evaluation and monitoring by multidimensional and multivariate statistical analysis becomes necessary to overcome the limits of simple descriptive and univariate statistics. In order to study the multivariate dependence problem of legal trial dura-tion in judicial flows, in this work we focus on a generalization of Partial Least Squares (PLS, Wold, 1966; Tenenhaus, 1998) called Multivariate Additive PLS via Splines (MAPLSS, Lombardo et al., 2007). MAPLSS permits to consider non-linearity and it is particularly opportune to capture interactions between variables of different nature, as the interaction itself, for example between the number of pending proceedings and the number of hearings, can affect the duration of trials more than the simple main effects (pending proceedings or number of hearing) separately do.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
