Approved For Release 2000/08/10 : CIA-RDP96-00792ROO0100120001-3 SECRET IAGPA-F-SD MEMORANDUM FOR RECORD 30 April 1984 SUBJECT: Trip Report, Stanford Research Institute (SRI), 4 & 5 March 1984 (U) 1. (U) Purpose: a. (S/CL-3/NOFORN) To discuss preliminary Data Base Management System (DBMS) requirements as they impact on Project CENTER LANE. b. (S/CL-3/NOFORN) To discuss the specifics of design in task directed DBMS programs which will support both training as well as operational parameters of Project CENTER LANE. C. (S/CL-3/NOFORN) To discuss the complexity and selected methodology of applied statistical analysis to be used to evaluate and support the DBMS programs developed in support of Project CENTER LANE. 2. (U) Problems: a. (S/CL-3/NOFORN) Data collected from current training and operational Psychoenergetic sessions while similar, require different DBMS handling during evaluation. b. (S/CL-3/NOFORN) Identifiable fields or characteristics which can be applied to the DBMS in both the training as well as the operational collection area range from a minimum of thirty (30) as defined by Dr. Jahns, or sixty (60) as tentatively identified by SRI- International (SRI-I) in support of training requirements. operational requirements could run as high as three or four hundred (300-400) dependent on the degree of detail in the final analysis. WARNING NOTICE: CENTER LANE SPECIAL ACCESS PROGRAM RESTRICT DISSEMINATION TO THOSE WITH VERIFIED ACCESS TO CATEGORY FOUR (4) SENSITIVE INTELLIGENCE SOURCES AND METHODS INVOLVED NOT RELEASABLE TO FOREIGN NATIONALS CLASSIFIED BY: CDR, INSCOM DECLASSIFY ON: OADR Approved For Release 206ECREDT96-00792ROO0100120001-3 Approved For Release 2000/08/10 : CIA-RDP96-00792ROO0100120001-3 SECR IAGPA-F-SD (30 April 1984) MEMORANDUM FOR RECORD SUBJECT: Trip Report, Stanford Research Institute (SRI), 4 & 5 March 1984 (U) C. (S/CL-3/NOFORN) The appropriate statistical methodology to be used in the evaluation of data collected in either the training or operational areas has not yet been determined. This is a sensitive issue since most scientific paradigms survive as effective singularly based on the accuracy of the statistical evaluation utilized to prove them. d. (S/CL-3/NOFORN) Lastly, a common data base between SRI-I and Project CENTER LANE presents no problem with respect towards the usual training session. However, this becomes complicated by an inability to share actual word data in the operational collection arena, due to its normal classification and sensitive nature when produced in support of an outside agency. 3. (U) Discussion Notes: a. (S/CL-3/NOFORN) Dr . Edward May has stated that he is certain a sufficiently tested and acceptable statistical analysis package can be written which will stand up to critic appraisal and comment. b. (S/CL-3/NOFORN) During discussions it was determined that a triad of empirical data bases could be designed which would be separate but provide intra-data base support to one another in performing the necessary functions to support both training and collection. (See Inclosure for a graphic representation of this concept.) Essentially these would be designed as follows: (1) (S/CL-3/NOFORN) First Data Base. This base would support training and contain the pre-determined information fields which would establish the psychoenergetic collector's statistical accuracy on a continuing basis throughout his/her training phases. It would break down the specific areas in which a given collector has significant strengths or weaknesses. It would serve to develop new fields which would be directly transferable to the operational collection or second data base area. Finally, it would provide the initial statistical accuracy figures for each trainee upon his/her graduation to operational collection. This information would be transferable to the third data base element. SECRET Approved For Release 2000/08/10 : CIA-RDP96-00792ROO0100120001-3 Approved For Release 20oo/SlEORF.9T-00792ROO0100120001-3 IAGPA-F-SD (30 April 1984) MEMORANDUM FOR RECORD SUBJECT: Trip Report, Stanford Research Institute (SRI), 4 & 5 March 1984 (U) (2) (S/CL-3/NOFORN) Second Data Base. Would contain all of the information from operational collection broken down into identifiable fields. These fields could be altered at will according to accuracy determination (s) based on feedback from the agency who both tasked the collection and has responsibility for assessing the original information provided. Information from this data base would be transferable to the third base in which statistical data on each operational psychoenergetic collector is maintained. However; in the development of brand new or unevaluated operational information, it would have the capacity of extracting the current statistical evaluation assumption for each collector from the third data base. (3) (S/CL-3/NOFORN) Third Data Base. Within the third data base would reside the on-going evaluations and statistical probabilities as pertains each individual psychoenergetic collector. These would be determined and automatically altered based on changes in the evaluated or known accuracy of information already logged in the second data base. These continual on-going re-evaluations of collector percentages and statistics which project current accuracy for any given collector would be utilized to predict the high and low areas of correctness for any given specific field in the second data base as new collection information is provided on a new tasking. C. (S/CL-1/NOFORN) As regards the inability to exchange actual word data the following design criteria would have to be utilized when designing the above data bases: (1) (S/CL-3/NOFORN) All word information fields would have to have two logic file areas. One for use by Project CENTER LANE, which would actually contain the word or phrase collected with the evaluated accuracy comment. One for use by SRI-I, which would have a numerical equivalent to the general field name and the associative accuracy comment. (See Inclosure two for a graphic illustration.) The program handling the CENTER LANE file area would have the capacity to translate to the numerical code prior to transfer of information to SRI-Ii b. (S/CL-1/NOFORN) Current code groups to identify specific tasked projects within Project CENTER LANE would remain the same and require no code translation. Approved For Release 20ooS/EGIF4&T6-00792ROO0100120001-3 Approved For Release 2000/08g.:-L[A-PP-9f~L-M792ROO0100120001-3 r-r-% 1- ki IAGPA-F-SD (30 April 1984) MEMORANDUM FOR RECORD SUBJECT: Trip Report, Stanford Research Institute (SRI), 4 & 5 March 1984 (U) C. (S/CL-1/NOFORN) Specific psychoenergetic collectors would be coded and handled as they have been since project concept. This list of collectors would be identifiable statistically to SRI-I, but only known by name within CENTER LANE. 4. (U) Additional Requirements. a. (S/CL-1/NOFORN) In the past proposals by SRI-I, they have suggested through proposal that CENTER LANE carry the cost or weight of purchase of computer equipment to support the above. Recommend that this not be the case and any cost of equipment which would revert to SRI-I upon successful completion of any contract for the above be born solely by them. flowever; it becomes apparent that because of such, the equipment purchased by them in support of the above may differ somewhat from that currently in use at CENTER LANE. Therefore, whatever interface requirement might have to be made in support of developing the above should probably be born by the CENTER LANE office. This should not be excessive. b. (S/CL-1/NOFORN) Design of the triad data base concept requires development of programming tree structures. SRI-I at this time probably does not possess the necessary main-frame capability to support an active continual run triad data base. This sh'ould not interfere however with design and testing of software development in support of CENTER LANE requirements. 5. (S/CL-I/NOFORN) Cost. There was no discussion of cost during this two day visit. SRI-I must generate a proposal for the above, which substantially deviates from their original DBMS proposal presented in January 1982. A detailed statement of work must be developed in conjunction with the CENTER LANE office prior to work commitment, and new performa of cost data must be provided by SRI-I prior to contracting. Suggest that such cost should not exceed $50K. (;0 ASB~_H!W. MCMONEA Automatic Data Processing Supervisor Project CENTER LANE Incl: 1 Triad Data Base Graph 2 Sub-Data Field Storage Example Approved For Release 2000/ QA,-RU_96-00792R0001 00120001-3 9EGRET Approved For Release 2000/08SEGRE-To792ROO0100120001-3 OPERATION OF TRAID DATE BASE INTRA-COMMUNICATIONS DATA BASE ONE TRAINING (Assists in Providing Newly Created Fields) DATA BASE THREE OPERATIONAL (Provides initial Statistical Accuracy) (Determines Initial Strength and Weaknesses) DATA BASE TWO OPERATIONAL (Updates Accuracy Based on Newly Evaluated Material) (Predicts New Target Accuracy) 10. Approved For Release 2000/08/ 96-00792ROO0100120001-3 Approved For Release 2000/08g E(; ffU792ROO0100120001-3 TRIAD DATA BASE GRAPH DATA BASE ONE DATA BASE TWO OPERATIONAL DATA BASE THREE OPERATIONAL PREDETERMINE FIELDS DEVELOP NEW FIELDS DETERMINE STRENGTHS DETERMINE WEAKNESSES1 STORE INITIAL OPERATIONAL INFORMATION STORE ACCURACY FEEDBACK PASS UPDATED ACCURACY EVALUATIONS STORE CURRENT STATISTICAL ACCURACIES MODIFY STATISTICAL ACCURACIES INSERT NEW ORIGINAL COLLECTOR ACCURACIES Approved For Release 200SEdCQ- 6-00792ROO0100120001-3 z Approved For Release 2000/081D MR' 792ROO0100120001-3 . L.URAE 0 SUB-DATA FIELD STORAGE PORTION OF DATA TREE EXAMPLE CENTER LANE INFORMATION EXAMPLE STATEMENT: "I see a fresh water lake adjacent to the target" Such a statement would be attributed in actual file to a specific collector and project number, along with an accurate or non-accurate notation. This may be recalled in the following format: "HU001-84-3, Robert Brown, Sub-file=Under Area, WATER=Accurate, FRESH=Inaccurate, LAKE=Accurate." "Estimated Accuracy under Sub-file AREA is 66%, WATER=100%, FRES11=00%, LAKE=100%." AREA WATER LAND FRESH OCEAN MOUNTAIN DESERT POIND I I 'GH 'UB LAKE COLD WARM HI LOW SCR SAND PORTION OF DATA TREE EXAMPLE SRI-I INFORMATION The same statement in the SRI-I file would riot be attributable to any specific project number but through use of a code number reflect the accuracy of a specific collector. Such as: "Collector=#009, Sub-file=Area, A/Sub-file 1=Accurate, B/Sub-file 1=Not Accurate, C/Sub-file I=Accurate." "Estimated Accuracy; Sub-file AREA is 66%, A/Sub-file=100%, B/Sub-file=00%, C/Sub-file=100%." AREA 1 2 11 1 3 4 1 2 3 4 6 7 Approved For Release 2000/QQV)-~W" 00792ROO0100120001-3 ET CAJ 1F\