Development and maintenance of automated crime investigation techniques

Frolov Vyacheslav Vladimirovich Lecturer of the Department of Criminology, Kuban State Agrarian University named after I.T. Trubilin 350044, Russia, Krasnodar Territory, Krasnodar, Kalinina str., 13, room 030 Slava.Frolov.84@mail.ru Other publications by this author

One of the factors that have a negative impact on the quality of the investigation of non-obvious crimes are the shortcomings and inaccuracies that investigators make when determining the subject of the investigation. The result is an incorrect choice of the direction of the investigation, an increase in the duration of the investigation, failure to establish the full extent of the circumstances to be proved in a criminal case, etc. Traditionally, one of the means to improve the formation of the subject of an upcoming investigation is the use by the investigator of scientific and practical recommendations of the investigation methodology of the appropriate type or group of crimes. Currently, there are sufficient opportunities not only to bring these forensic knowledge to investigators in writing, but also to create an automated crime investigation methodology – a type of automated information system (hereinafter – AIS) to support decision-making.We partially agree with the opinion that at the moment the proposed automated investigation methods are only private investigation methods implemented in the form of electronic textbooks, textbooks, multimedia tools, i.e. the field of forensic didactics (improving pedagogical methods of teaching criminology, improving the skills of investigators and interrogators) ^{[1, pp. 3-8]}. But let us disagree with the statement that automation of investigation techniques for certain types of crimes is nothing more than ensuring greater accessibility of forensic knowledge, simplifying the search for scientific and forensic data in scientific and methodological sources using modern computer technologies ^{[1, pp. 3-8]}.

The creation of AIS to support the investigator's decision-making in the investigation of crimes has already taken place (the information and reference software package "Automated workplace of the investigator" was developed by the Research Institute of the Academy of the Prosecutor General's Office of the Russian Federation with the participation of the St. Petersburg Law Institute (branch) Academy of the Prosecutor General's Office of the Russian Federation ^[2]; mobile application "CrimLib.info – Investigator's Handbook" for iOS devices ^[3], etc.). In our opinion, they did not find wide practical application primarily due to the design (technical) features of their functioning, which significantly reduced the effectiveness of their use. Firstly, since the database in this program was based on typical investigative situations and standard versions, the investigator needed to determine the investigative situation of the investigation being conducted, compare it with the proposed typical ones and select the appropriate one. After that, he used the system of standard versions proposed for this situation, as well as algorithms for procedural, investigative, organizational and preparatory actions and operational investigative measures. At the same time, the program did not provide for the possibility of introducing information by the investigator obtained as a result of the investigation of a crime, both as part of the specification of standard versions, and as part of making changes to the system of investigation goals and the algorithm of proposed actions. Which, in fact, could not contribute in any way to the quality of the investigation, since such planning principles as individuality and dynamism were violated. Secondly, the program did not provide for the formation of a new database on the crime being investigated by the investigator, which would allow using the information accumulated there to compile procedural (protocols, resolutions, indictment) and other auxiliary (optional) documents. Thirdly, the program proposed a direction of investigation, to a greater extent determined by the investigative situation, but did not contain systems of logical consequences that would be derived from standard versions, i.e. it did not fully reflect the mechanism of targeting.

Therefore, it is necessary to determine the structural components (databases) of the automated investigation methodology, the principles of functioning of its algorithms, etc., which will allow the development and effective use of AIS data in the investigation.

Using the example of the investigation of traffic crimes, we will try to consider this problem and propose ways to solve it. About 24% of investigations of road traffic crimes are complex and particularly complex (there are no eyewitness witnesses to the crime, the driver involved in the accident is unknown and fled the scene, etc.). In such cases, the process of forming a system of circumstances to be established in a criminal case is also complicated. The investigator's use of only the method of specifying the subject of proof in forming a system of investigation goals will not be enough, since a significant part of the circumstances of the accident is unknown, and therefore can only be expressed by a certain number of different variants of their existence. Concretization is possible only when most of the circumstances to be established are confirmed at the beginning of the investigation by known direct evidence ^{[4, pp. 51-52]}. A unified position has been formed in forensic science that the main method of cognition of an unobvious crime that has occurred is the construction and verification of versions ^{[5, p. 29; 6, p. 49]}. At the same time, the investigative version has two more functional meanings: tactical and organizational ^{[7, p. 345]}. The problem of the role of the version in the planning of an investigation was first raised in his works by V. I. Gromov, calling the version the logical basis of planning ^{[8, p. 80]}. Subsequently, this issue was investigated by S. A. Golunsky, B. M. Shaver, T. M. Arzumanyan, A. N. Vasiliev, G. N. Alexandrov, A. N. Kolesnichenko, A. R. Shlyakhov, I. M. Luzgin, G. V. Artsishevsky, and others. R. S. Belkin indicates that one of the elements of planning is the nomination of versions and the definition of the tasks of the investigation ^{[7, p. 381]}. In general, sharing this point of view of R. S. Belkin, let us note that in the structure of the organization, as V. D. Zelensky precisely pointed out ^{[9, p. 55]}, the construction of versions has an independent organizational significance as an element of target determination. Determining the objectives of the investigation is not limited only to analyzing the version, it is a complex intellectual activity that requires the investigator to take into account such circumstances as the specifics of the crime under investigation, its criminal law and criminalistic characteristics, etc. Targeting precedes planning.

Thus, when investigating non-obvious crimes against traffic safety and operation of transport, the specification of unknown circumstances of the crime (the person who committed the crime, the mechanism of the accident, etc.) is carried out by building versions. The analysis of versions allows the investigator to determine the goals of the upcoming activity. The targeting mechanism in this case looks like this:

– the maximum possible number of logical consequences is derived from each version. A logical consequence is a circumstance that may exist if the version is correct (if → then);

– the whole set of logical consequences derived from the version is a set of intermediate facts (indirect evidence) that make it possible to substantiate (certify) the circumstances of the subject of proof;

– the establishment (confirmation or refutation) of all intermediate facts are the tactical goals of the investigation, i.e. the goals of investigative actions, since this is how version verification is carried out: confirmation in the version of all logical consequences is the transition from a probable assumption to reliable knowledge.

From the perspective of the algebra of logic (a section of mathematical logic that studies the structure (form, structure) of complex logical statements and ways to establish their truth using algebraic methods ^{[10, p. 137]}) – the version (condition) and the logical consequence derived from it are two statements connected by a bundle "if... then" that is, a logical operation is an implication, while only the truth can follow from the truth, and since the version is an assumption about the truth, therefore, confirmation of the truth of the consequences form the truth of the condition (version). Logical operations, in turn, are the basis for the functioning of a computing system, i.e. the same computer devices can be used to process and store numerical information presented: a) in a binary number system; b) in logical variables ^[11]. Thus, we have the opportunity to implement the process of determining the objectives of an investigation in information technology, by converting it into electronic and digital form with implementation in the form of a computer program (AIS). The fact that the most promising direction of using computer technologies to ensure the investigation of crimes is the creation of AIS has been noted by many criminologists, and most of them, as a central component of such a system, determine the element of planning the investigation of crimes ^[12]. The structuring of the AIS elements must be carried out on the basis of the theoretical provisions of criminalistics, in addition, it is necessary to take into account the norms of the Code of Criminal Procedure of the Russian Federation regulating criminal proceedings. Sharing the position that the definition of the objectives of the investigation precedes planning, is not limited only to the construction and analysis of versions, forms the structure of the investigation, we believe that the foundation (basis) The AIS should be a database (hereinafter referred to as the database) "Targeting". This database should include two blocks:

1) information on the circumstances to be established and proved during the investigation of traffic crimes. Moreover, the investigator should be able to enter data that allows specifying these circumstances, guided by the actual data of the investigation he is conducting. At the same time, the system of goals formed according to the crime under investigation is fixed in the AIS as a separate file, with the possibility for the investigator to access its contents and adjust the system of goals taking into account the progress of the investigation;

2) this block provides the process of targeting by building investigative versions. The content and structure of this block, in our opinion, is determined by the following circumstances:

– the place of organization of crime investigation in the system of criminology and its relationship with the methodology of crime investigation. Scientific discussion on this issue has been conducted since the beginning of the investigation of the organization of the investigation ^[13]. Gradually, the point of view was formed that the organization of the investigation should be considered in the general theoretical part of criminology ^{[14, pp. 47-50]}. At the moment, the dominant point of view is that the organization of an investigation is a forensic theory, part of the general theory of criminology, the subject of which is the regularities of ordering an investigation, the object is the activity of creating an investigation structure and optimal conditions for its production;

– the criminalistic methodology develops a strategy for all criminalistic activities to investigate crimes ^{[15, p. 74]}. The subject of this activity is the circumstances to be proved and established during the investigation of a specific type or group of crimes, i.e. the subject of the investigation, which in turn determines the means of its establishment (investigative actions and their combination), as well as the tactical component of their content. Thus, the methodology represents a program of activities for the investigation of a type or group of crimes, which allows the investigator to create a model of a specific investigation being carried out. Therefore, the task of the investigation methodology is to assist the investigator in investigating a specific crime, reasonably ensuring that the correct forensic, procedural, organizational and tactical decisions are made;

– the organization of an investigation is the creation of an investigation structure, the forensic methodology contributes to the concretization of this structure.

Thus, this block, taking into account the circumstances considered, should be formed based on general (provisions of the organization of the investigation) and particular (provisions of the criminalistic methodology for investigating a type or group of crimes, in our case traffic crimes) patterns. According to the general opinion, the constituent elements of the structure of the criminalistic methodology for investigating crimes are typical investigative situations that arise at various stages of the investigation, versions ^{[15, p. 198]}. Typical investigative situations and standard versions are an integral part of the investigation methodology of the type and group of crimes through which the program of the upcoming investigation is formed. The problems of typical investigative situations and typical versions have been studied quite fully and in detail in criminology, in practice, programs (algorithms) developed on their basis for investigating certain types and groups of crimes are widely used ^{[16, p. 383]}. The possibility of creating systems of logical consequences was considered by criminologists. Thus, A. S. Zhiryaev proposed the division of intermediate facts into general and particular (special) ones. The former take place in the investigation of all crimes, the latter characterize one or another kind or type of illegal acts ^{[17, pp. 62-63.]}. A. A. Khmyrov pointed out that the development of lists of intermediate facts (sets of evidence) for typical investigative situations is an important area of investigation methodology and should be carried out in two directions: the range of intermediate facts for general investigative situations, without taking into account the type of crimes, and a system of intermediate facts for specific investigative situations characteristic of the investigation of certain types or groups of crimes ^{[18, p. 231]}. It should be noted that the development of systems of logical consequences by criminologists was of a one-time nature, there were practically no comprehensive studies in this direction. In this context, we note the work of N. V. Ageev, in which an attempt was made to systematize logical consequences by groups for certain types of crimes (murder, rape, theft, robbery, robbery) ^{[19, pp. 69-73]}. So, in the investigation of non-obvious murders, the following groups of logical consequences are proposed: 1) aimed at establishing the fact of finding a person suspected of murder on the spot and during its commission; 2) aimed at establishing the fact of preparation for murder (selection, manufacture, etc. of the murder weapon, other means used in its commission); 3) aimed at establishing the fact of finding traces on things, the body of the suspect murders; 4) aimed at establishing the suspect's motive and intent of murder ^{[19, pp. 69-70]}. The group principle of logical consequences, which is the basis, in our opinion, allows us to obtain a system of logical consequences (intermediate facts), but to use this system in AIS, it is necessary to specify logical consequences, since the effectiveness of information technology is determined primarily by providing complete, clearly conceptually formulated information. It is this approach to the formation of the database "Targeting" that will make a significant breakthrough in the creation of automated investigation techniques as decision support systems, and in the future, the training of this system will ensure its genesis to the level of an information intelligent system.

Initially, it is impossible to provide for all possible specific investigation situations in this automated system, in addition, at this stage it is not necessary to create it. Firstly, this system does not make a decision for the investigator, but simplifies his intellectual and analytical activities to create a structure for the upcoming investigation. The goals and means proposed by the system to achieve them will indeed be an approximate plan for a specific investigation carried out by the investigator, but this plan will be drawn up in compliance with scientific and practical forensic recommendations. The specification of this plan within the framework of the ongoing investigation, as we have already indicated, is an integral part of the investigator's own activities. Secondly, work on an automated investigation methodology is an ongoing process, it does not end at the stage of its creation and implementation into investigative practice. Its effective application is impossible without constant updating of the database. Updating databases with reliable, complete, scientifically based, criminally significant information, in fact, will be the basic criterion for the functioning of this automated system. The accumulation in the database of significant amounts of such information obtained, including as a result of the analysis of completed investigations of crimes of this type or group, will allow it to increase the number of variants of typical investigative situations, standard versions, systems of intermediate facts, algorithms of the investigator's actions. The technical capabilities of an automated system (a large amount of information stored in its memory; the speed of its processing, etc.) make it possible to get away from the tendency in forensic science to avoid excessive detailing of investigative situations in theory. Since, in practical terms, the multiplicity of situations leads to the creation of many complex recommendation systems, which will negatively affect the quality of the investigation ^{[20, p. 100]}. Thus, in an automated system, a multiplicity of situations and a multitude of systems of recommendations for them will allow the investigator to quickly choose the one that is most similar to the investigative situation of the investigation he is conducting. This result can be achieved only by continuing after the commissioning of the training of an automated system for the investigation of a certain type or group of crimes. In this case, training is conditioned in two ways: on the one hand, by the patterns of creation and effective functioning of intelligent information systems (in our vision, in the process of genesis, with proper implementation of training, an automated method for investigating a certain type or group of crimes from the category of decision support systems to the category of information intelligent systems (IT systems)), and on the other - the task which needs to be solved within the framework of automating the methodology of crime investigation.

In conclusion, we formulate a number of generalizations:

1. It is necessary to really start developing automated methods for investigating crimes and introducing them into investigative activities. This is a difficult and time-consuming job, it can only be carried out by creating working groups, which must necessarily include both forensic scientists and specialists in the field of information technology. The quality and completeness of the information entered into this system is undoubtedly the responsibility of criminologists. In addition, the algorithms for the functioning of the system should be based on the provisions and patterns of forensic science.

2. The foundation (foundation) The AIS should be a database "Targeting", which includes two blocks: 1) information about the circumstances to be established and proven in the investigation of traffic crimes; 2) a block that ensures the formation of goals through the construction of investigative versions.

3. The system of goals formed by the AIS is fixed in a separate file, with the ability of the investigator to access its contents in order to adjust it taking into account the progress of the investigation. The investigation plan proposed by AIS will be an approximate one, formed taking into account the circumstances of the crime under investigation and in compliance with scientific and practical criminalistic recommendations for the investigation of this type or group of crimes. The specification is made directly by the investigator.

4. In our opinion, these measures will make it possible to make a significant breakthrough in the creation of automated investigation techniques as decision support systems, and in the future, the training of this system will ensure its genesis to the level of an information intelligent system. In fact, a positive solution to this issue can be implemented at the fourth level of the organization of crime investigation – the organization and management of crime investigation in the country.