Linguistic Evidence in Security, Law and Intelligence

https://lesli-journal.library.pitt.edu/ojs/lesli/issue/feed Linguistic Evidence in Security, Law and Intelligence 2019-09-13T08:00:13+00:00 Carole E. Chaski PhD lesli@mail.pitt.edu Open Journal Systems Who wrote it? Who said it? Is this text similar to that one? Can we group those texts together? Is this text really what it purports to be? What can we know about the kind of person who generated this text --age, dialect, native language?These are the kinds of questions that can turn an investigation around, and each one relies on language as evidence. Linguistic evidence can and has already played a crucial role in homeland security, counter-terrorism, criminal and civil investigations, national intelligence, business intelligence and executive protection.LESLI is an interdisciplinary journal for linguists, computer scientists, psychologists, psychiatrists, attorneys, law enforcement, security executives, and intelligence analysts. As the journal of the Institute for Linguistic Evidence and its membership TALE: The Association for Linguistic Evidence, LESLI provides a forum to present rigorous research, requests for research, and policy discussions.READERSUse the <a href="/ojs/index.php/lesli/user/register">Register</a> link at the top of this page. As a registered reader, you will receive the Table of Contents by email for each new issue of the journal. See the journal's <a href="/ojs/index.php/lesli/about/submissions#privacyStatement">Privacy Statement</a>, so you that know that your name and email address will not be used for other purposes. Thank you in advance for registering as registration allows the journal to keep track of your support and total readership.AUTHORSInterested in submitting to this journal? We recommend that you review the <a href="/ojs/index.php/lesli/about">About the Journal</a> page for the journal's section policies, as well as the <a href="http://lesli-journal.library.pitt.edu/ojs/index.php/lesli/about/submissions#authorGuidelines">Author Guidelines</a>. Authors need to <a href="/ojs/index.php/lesli/user/register">register</a> with the journal prior to submitting.When you register, be sure that you check Author role (Reader is checked by default). If you do not have Author checked you will not be able to submit a manuscript online. If you have already registered, and you cannot find a place to submit, then go to your profile and make sure that the Author role is checked.If you have already registered and your profile includes the Author role, you can can simply <a href="/ojs/index.php/index/login">log in</a> and begin the five-step submission process.LIBRARIANS AND CYBRARIANSThank you in advance for listing this journal among your library's electronic journal holdings. We highly recommend the University of Pittsburgh ULS as our online publisher. LESLI's open source publishing system is suitable for libraries to host for your faculty and corporate members to use with journals they are involved in editing (see <a href="http://pkp.sfu.ca/ojs">Open Journal Systems</a>). https://lesli-journal.library.pitt.edu/ojs/lesli/article/view/19 Developing and Analyzing a Spanish Corpus for Forensic Purposes 2019-09-13T08:00:13+00:00 Ángela Almela angelalm@um.es Gema Alcaraz-Mármol gema.alcaraz@uclm.es Arancha García-Pinar arancha.garcia@upct.es Clara Pallejá clara.palleja@cud.upct.es

In this paper, the methods for developing a database of Spanish writing that can be used for forensic linguistic research are presented, including our data collection procedures. Specifically, the main instrument used for data collection has been translated into Spanish and adapted from Chaski (2001). It consists of ten tasks, by means of which the subjects are asked to write formal and informal texts about different topics. To date, 93 undergraduates from Spanish universities have already participated in the study and prisoners convicted of gender-based abuse have participated. A twofold analysis has been performed, since the data collected have been approached from a semantic and a morphosyntactic perspective. Regarding the semantic analysis, psycholinguistic categories have been used, many of them taken from the LIWC dictionary (Pennebaker et al., 2001). In order to obtain a more comprehensive depiction of the linguistic data, some other ad-hoc categories have been created, based on the corpus itself, using a double-check method for their validation so as to ensure inter-rater reliability. Furthermore, as regards morphosyntactic analysis, the natural language processing tool ALIAS TATTLER is being developed for Spanish. Results shows that is it possible to differentiate non-abusers from abusers with strong accuracy based on linguistic features.

2019-09-13T00:00:00+00:00 Copyright (c) 2019 Linguistic Evidence in Security, Law and Intelligence https://lesli-journal.library.pitt.edu/ojs/lesli/article/view/20 Benchmarking Author Recognition Systems for Forensic Application 2019-09-13T08:00:13+00:00 Hans van Halteren b.v.halteren@let.ru.nl

This paper demonstrates how an author recognition system could be benchmarked, as a prerequisite for admission in court. The system used in the demonstration is the FEDERALES system, and the experimental data used were taken from the British National Corpus. The system was given several tasks, namely attributing a text sample to a specific text, verifying that a text sample was taken from a specific text, and verifying that a text sample was produced by a specific author. For the former two tasks, 1,099 texts with at least 10,000 words were used; for the latter 1,366 texts with known authors, which were verified against models for the 28 known authors for whom there were three or more texts. The experimental tasks were performed with different sampling methods (sequential samples or samples of concatenated random sentences), different sample sizes (1,000, 500, 250 or 125 words), varying amounts of training material (between 2 and 20 samples) and varying amounts of test material (1 or 3 samples). Under the best conditions, the system performed very well: with 7 training and 3 test samples of 1,000 words of randomly selected sentences, text attribution had an equal error rate of 0.06% and text verification an equal error rate of 1.3%; with 20 training and 3 test samples of 1,000 words of randomly selected sentences, author verification had an equal error rate of 7.5%. Under the worst conditions, with 2 training and 1 test sample of 125 words of sequential text, equal error rates for text attribution and text verification were 26.6% and 42.2%, and author verification did not perform better than chance. Furthermore, the quality degradation curves with slowly worsening conditions were not smooth, but contained steep drops. All in all, the results show the importance of having a benchmark which is as similar as possible to the actual court material for which the system is to be used, since the measured system quality differed greatly between evaluation scenarios and system degradation could not be predicted easily on the basis of the chosen scenario parameters.