ALIAS Developed by Dr. Carole Chaski, ALIAS is our core software product we use to analyze linguistic evidence at the Linguistic Evidence Laboratory and through web_ALIAS (a web-accessed version to which trained users can subscribe). ALIAS is: Automated: We automate our methods through computer programs so that our methods are objective, not undermined by human fatigue, and process as fast as we can possibly make them. Our computer programs analyze each case document in the same way, whether the document is being questioned or used as a comparison document. Thus, there is no subjectivity in analysis which can lead to confirmation bias. Automating analysis through computer programs guards against confirmation bias and leads to objective and replicable results. Linguistic: Our methods are built on fundamental principles of modern linguistic theory. We are trained as linguists and rely on standard linguistic methods. We apply current knowledge from linguistics to solve forensic problems. We use computational linguistics to implement the application of linguistic theory and methods to forensic problems. Identification: Among forensic problems for language evidence, author identification is a central issue. Our classification methods focus on identifying authorship through comparing known documents to any questioned document(s). Assessment: Other important problems for language evidence involve assessing a document for its investigative value. Our assessment methods focus on determining the authenticity of threats and suicide notes, the similarity of trademarks and service marks, linguistic profiling for gender, age, educational level, non-native English and veracity, and the relationship between documents. System: ALIASsm is a system which includes a central database of documents, analytical components which quantify relevant linguistic patterns, statistical components which analyze the quantification output, and reporting components which present the results. ALIAS modules answer forensically significant questions, e.g., who wrote it? is this a real suicide note? is this a real threat letter? and so forth. Each ALIAS module accesses the document database, analyzes the document extracting and counting specific linguistic patterns, implements a statistical analysis of the pattern counts, and reports an answer. Some ALIAS modules are totally automated and can be accessed through web_ALIAS, by trained users. Trainees must meet strict qualifications and pass examinations to become users. Other ALIAS modules are semi-automated and require a trained and degreed linguist to monitor automated pattern checking and thus can only be used in ALIAS Technology's Linguistic Evidence Laboratory.

ALIAS

Developed by Dr. Carole Chaski, ALIAS is our core software product we use to analyze linguistic evidence at the Linguistic Evidence Laboratory and through web_ALIAS (a web-accessed version to which trained users can subscribe). ALIAS is:

Automated: We automate our methods through computer programs so that our methods are objective, not undermined by human fatigue, and process as fast as we can possibly make them. Our computer programs analyze each case document in the same way, whether the document is being questioned or used as a comparison document. Thus, there is no subjectivity in analysis which can lead to confirmation bias. Automating analysis through computer programs guards against confirmation bias and leads to objective and replicable results.

Linguistic: Our methods are built on fundamental principles of modern linguistic theory. We are trained as linguists and rely on standard linguistic methods. We apply current knowledge from linguistics to solve forensic problems. We use computational linguistics to implement the application of linguistic theory and methods to forensic problems.

Identification: Among forensic problems for language evidence, author identification is a central issue. Our classification methods focus on identifying authorship through comparing known documents to any questioned document(s).

Assessment: Other important problems for language evidence involve assessing a document for its investigative value. Our assessment methods focus on determining the authenticity of threats and suicide notes, the similarity of trademarks and service marks, linguistic profiling for gender, age, educational level, non-native English and veracity, and the relationship between documents.

System: ALIASsm is a system which includes a central database of documents, analytical components which quantify relevant linguistic patterns, statistical components which analyze the quantification output, and reporting components which present the results. ALIAS modules answer forensically significant questions, e.g., who wrote it? is this a real suicide note? is this a real threat letter? and so forth. Each ALIAS module accesses the document database, analyzes the document extracting and counting specific linguistic patterns, implements a statistical analysis of the pattern counts, and reports an answer.

Some ALIAS modules are totally automated and can be accessed through web_ALIAS, by trained users. Trainees must meet strict qualifications and pass examinations to become users. Other ALIAS modules are semi-automated and require a trained and degreed linguist to monitor automated pattern checking and thus can only be used in ALIAS Technology's Linguistic Evidence Laboratory.

Handling Language as Evidence Responsibly, Respectfully and Reliably

In some criminal, civil, and security matters, language itself can be evidence. A suicide note, a threatening letter, anonymous communications, business emails, blog posts, trademarks --all of these can help police detectives, private investigators, prosecutors, defense attorneys, plaintiff attorneys, human resource executives and private individuals understand the heart of an incident. When you are faced with a suspicious document, whether you need to know who wrote it, or if it is a real threat or a real suicide note, or if it is too close for comfort to some other document, ALIAS Technology LLC can help you.

We provide a unique service of using forensically-feasible computational tools for handling language as evidence. ALIAS --Automated Linguistic Identification and Assessment System-- is the main software we use to answer forensically-relevant questions in an objective, repeatable, linguistically-acceptable way.

Based on ALIAS, sworn testimony regarding authorship identification has been deemed fully admissible in both Federal and State Courts under both the Daubert and Frye standards for scientific evidence. ALIAS has been admitted after two Daubert hearings (one Federal, one State) and one Frye hearing (State). But most importantly, our service

has helped clients get to the truth of a matter and determine their best strategy in a difficult situation.

The Institute for Linguistic Evidence (ILE) is our sister, the first non-profit research organization devoted to developing and validating linguistic methods. ILE functions as the research and development wing of ALIAS Technology LLC. Click here to visit the ILE website.

Why Does ALIAS Work?

Since 1992, Dr. Chaski has taken into account the fact that language can be imitated. We all know someone who can mimick another person's voice with uncanny similarity, and we know that a literary artist might purposefully pen a story in the style of a famous antecedent. Forgery, however, is only as good as the forger. What makes a linguistic forger hard to detect is metalinguistic awareness, or the ability to think about language as an object in and of itself, and not merely as a means of conveying a message. Each human has some degree of metalinguistic awareness, and while some people have an enormous aptitude for linguistic mimickry based on metalinguistic awareness, two psycholinguistic facts enable ALIAS to work.

First, since language is built for communication, or the what of a message, very little attention is paid to the structure of language, or the how of a message. In fact, psycholinguistic experiments since the 1960's have shown that the how of a message degrades from memory in milliseconds; witness the fact that we use court reporters to transcribe how something is said because we simply don't remember the exact words for even a minute. These kinds of hard-to-remember linguistic features are the ones on which Dr. Chaski has focused her research and development underlying ALIAS, and consequently ALIAS modules perform very well on their particular tasks.

Second, since people vary in metalinguistic awareness as in intelligence and other cognitive characteristics, only a small percentage of the population will have the kind of metalinguistic awareness necessary for linguistic forgery. It takes hardly any metalinguistic awareness to copy someone's punctuation marks or frequent words or spelling errors --the kinds of linguistic features used by forensic stylistics, text analysis, forensic sociolinguistics or behavioral analysis. But it takes a much more sophisticated, and much more rare, metalinguistic awareness to imitate syntactic structures at exactly the right proportions to withstand statistical analysis which differentiates different linguistic classes.

In sum, ALIAS works because it uses sophisticated and normally unconscious linguistic features which are very difficult for us to recognize, manipulate and control. While linguistic forgery is possible, and each ALIAS component's error rate is determined by continual validation testing, analysis by ALIAS can detect linguistic forgery and reliably classify documents for authorship and text type.

ALIAS and Legal Standards for Admissible Evidence

We respect the legal standards for admissible expert scientific evidence as put forth in rulings of the United States Supreme Court, Federal Rules of Evidence and State Rules of Evidence. In particular, ALIAS methods are designed to meet both the Frye and Daubert standards for admissible evidence. The technologies underlying ALIAS are generally accepted within the general field of linguistics and the specific field of computational linguistics. ALIAS modules are not released from testing deployment into consulting deployment until they reach at least 80% accuracy, and most have smaller error rates. ALIAS reports include the current error rates from validation testing so that investigators, juries and judges can weigh the results appropriately.

Each ALIAS module has been:

Developed using standard methods in the academic and industrial field of computational linguistics (also known as natural language processing or natural language engineering).

Computational linguistics is an increasingly important subfield in linguistics. In computational linguistics, linguists develop software which enables computers to perform linguistic behaviors: the most famous example is the Internet search engine, which is a computer program imitating a human's ability to search the language of documents and classify documents by topic. The core text analysis modules of ALIAS currently employ a 30,000+ word lexicon, morphological rules, and syntactic rules to tag words for part of speech and categorize words semantically.

Designed to combine linguistic analysis with statistical processing.

The statistical procedures are standard procedures in the academic and industrial fields of statistics, machine learning and data mining. Some ALIAS modules output linguistic pattern analysis and quantification for processing with statistical software such as IBM's SPSS, Sherrod's DTREG, or the open source Weka program. Other ALIAS modules implement the statistical analysis directly based on models developed in SPSS.

Tested on suitable databases of documents appropriate for the forensic question.

In total, the ALIAS Documents Library currently contains pertinent document types organized into relevant corpora which are used to develop and test ALIAS modules. These documents have been gathered from print and electronic published sources, paid subjects in funded data gathering, FOIA requests, voluntary donations, and cases we have worked. From the ALIAS Documents Library, we create the development and testing corpus specific to each module. In development and testing, we only use documents for which we have known provenance and other information essential to the module; that is, for example, when we develop and test methods for authorship identification modules, we only use documents in our testing corpus for which we have known authorship, and so forth. Due to the fact that we can not have the highest level of information regarding provenance of all electronic sources, we can not use all of the ALIAS Documents Library for each module's development and testing. We use the largest possible number of documents to create a relevant corpus, and we constantly expand the core ALIAS Documents Library. If you wish to donate documents, please contact Dr. Chaski at cchaski@linguisticevidence.org .

Access to ALIAS

Due to the sensitive nature of ALIAS functions, access to ALIAS is restricted to trained users currently serving as law enforcement, licensed private investigators and security consultants, attorneys, and corporate officers in human resources, executive protection, and business intelligence.

Language as Evidence ALIAS Services Multimedia/Press/Publications Book Store Corporate