ICS211 | Introduction to Computer Science | Boyer Moore's Algorithm

The Boyer-Moore algorithm is associated with the usage of information that are generally collected at the preprocess phase for the purpose of skipping certain sections of the text and this initially results in obtaining a constant factor which is lower than most of the string search algorithms. Particularly it is better to state that this algorithm runs at a much faster rate and the major reason for this is that the length of the pattern increases gradually (Rahim et al., 2017). Besides this some of the most important features of this algorithm includes the matching at the patterns tail instead of matching at the head. Another feature of this includes the skipping along the text by taking a multiple character jump instead of searching each and every character present in the text.

Explanation of the Algorithm:

Due to all this reason the Boyer-Moore algorithm can also be defined as the most efficient string-matching algorithm which is being used by the usual applications. One of the simplified versions of this algorithm or the most simple version of the entire algorithm. This is often implemented in text editors in order use some commands like the «search» and «substitute» command.

This algorithm is associated with the scanning of certain characters that are present in the pattern from right to left and this generally begins from the one present at the extreme right (Angeli et al., 2015). Whenever a mismatch is detected, it makes use of two functions which precomputed for the purpose of shifting the window to the right. There are two type of shift functions and these two shift functions are namely the good-suffix shift (also known as the matching shift and the bad-character shift (also known as the occurrence shift).

Bad character Heuristic: The idea of bad character heuristic is very simple. Bad characters are those characters which does not match with the characters of the text (Jaiswal, 2014). Whenever a mismatch occurs shifting of the pattern is done unless and until the mismatch matches and the pattern P moves past the character which is mismatching.

Case 1: Mismatch becomes a match;

The last place of mismatch occurrence is identified in the pattern provided and in case if the mismatch character exists in the pattern the shifting of the pattern is done in such a way that it gets aligned with the mismatching character in the text.

Case 2: When the pattern moves past the mismatching character:

In this type of cases the position of the last mismatching character in the pattern is identified and in case when the character is not existing then, shifting of the pattern is done past the mismatching character.

Good suffix Heuristics: this is another variation of the Boyer-Moore algorithm and just like the bad character heuristic the good character heuristic is also associated with the generation of a pre-processing table. Let t be the substring of the text T which is to be matched with the substring of another pattern let it be P (Jeong et al., 2015). followed by this shifting of the pattern is done until three major criteria are met and this criterion include the following.

• Another occurrence of the t in the P matched with the t in the T
• A prefix of the P, which matches with the suffix of the t
• P moves past the t.

Type of Bias occurring in the hiring process:

• Confirmation bias: This the type of bias which particularly occurs when peoples are associated with the creation of a hypothesis in their mind and are looking for ways so as to prove it. This is considered to be an innate tendency which seeks out for confirmation of the preconceived beliefs.
• Effective Heuristics: This term can be considered to be bit technical, however the meaning lying behind this is very much simple. This type of bias generally happens whenever an interviewer is associated with judging an individual based upon the superficial factors which includes the visible tattoos or personal body weight standards and many more (Domínguez, Carballo & Núñez, 2017). In this case the interviewer might be associated with taking decisions depending upon the one-dimensional characteristics and not upon the important ones like the problem-solving skills and many more.
• Expectation anchor: this happens when an interviewer is associated with bypassing the proper investigation regarding the background of the candidate and depends on the frivolous expectations of the anchors thereby leading to the favouritism of a candidate. The interviewer is associated with believing the fact that the candidate is more suitable for the job than the others which initially puts a mental block for the interviewer during the other interview which is going to be conducted later (Waga, Akazaki & Hasuo, 2016).
• Intuition: in case if an interviewer is associated with making of decisions upon the sixth sense of an individual then the interviewer is intuitively selecting the candidate or is intuitively rejecting all the other candidates. This generally happens when the candidate’s information is readily available on hand.

Boyer-Moore Algorithm in Hiring process:

“bowmo”, is an evolutionary software as a service (SaaS) which has entered the HR technology in order to eliminate the bias occurring in the process hiring. Elimination of the bias is generally done for the database and during the process of searching the resumes (Jeong et al., 2015). This is the software which has is associated with enabling the recruiters and the hiring managers to search for candidates having specific skills according the job requirements.

This is technology which makes use of the Boyer-Moore string search algorithm and besides this by using the concept of this algorithm the technology helps in eliminating the Human bias. The Boyer-Moore string search algorithm is a benchmark for modern search engines like Google. bowmo is the software which is associated with finding the right candidates so as to empower the Hiring Managers to keep outside and corporate recruiters accountable.

Along with removing of the bias the software “bowmo is associated with enabling the companies in building up of various kind of talent pools as well this is the software which ca be integrated very easily and is responsible for seamless matching of the qualified employees with the jobs which are appropriate for them (Nasution et al., 2018). Usage of this also makes the recruiters have a clear view without getting stuck due to the information which are responsible for causing distraction and are irrelevant.

Conclusion:

The report helps in understanding the fact that the algorithm is not associated with considering certain factors or using certain factors for the matching the fields and these factors include the gender, caste, color, race, name sexual orientation or religion. The data related to EEOC are collected by making use of different systems or applications and this data are not the data fields required for matching a particular job. Some of the most important Credentials includes the role that has been played, years of experience, the technical skill and sometimes education acts as the few of the most relevant criterions. bowmo is one of the software using the Boyer-Moore String Search Algorithm that is associated with combatting the bias and making this AI and recruiting convergence a reality.

References:

Angeli, P., Ginès, P., Wong, F., Bernardi, M., Boyer, T. D., Gerbes, A., ... & Moore, K. (2015). Diagnosis and management of acute kidney injury in patients with cirrhosis: revised consensus recommendations of the International Club of Ascites. Gut, gutjnl-2014.

Boyer, R. S., & Moore, J. S. (2014). A computational logic handbook: Formerly notes and reports in computer science and applied mathematics. Elsevier.

Domínguez, A., Carballo, P. P., & Núñez, A. (2017, July). Programmable SoC platform for deep packet inspection using enhanced Boyer-Moore algorithm. In Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC), 2017 12th International Symposium on (pp. 1-8). IEEE.

Jaiswal, M. (2014). Accelerating Enhanced Boyer-Moore String Matching Algorithm on Multicore GPU for Network Security. International Journal of Computer Applications, 97(1), 30-35.

Jeong, Y., Lee, M., Nam, D., Kim, J. S., & Hwang, S. (2015). High performance parallelization of Boyer–Moore algorithm on many-core accelerators. Cluster Computing, 18(3), 1087-1098.

Jeong, Y., Tran, N. P., Lee, M., Nam, D., Kim, J. S., & Hwang, S. (2015). Parallelization and Performance Optimization of the Boyer-Moore Algorithm on GPU. KIISE Transactions on Computing Practices, 21(2), 138-143.

Nasution, M. D. T. P., Rossanty, Y., Achmad Daengs, G. S., Sahat, S., Rosmawati, R., Kurniasih, N., ... & Kadir, I. A. (2018). Decision support rating system with Analytical Hierarchy Process method. Int. J. Eng. Technol, 7(2.3), 105-108.

Rahim, R., Ahmar, A. S., Ardyanti, A. P., & Nofriansyah, D. (2017, December). Visual Approach of Searching Process using Boyer-Moore Algorithm. In Journal of Physics: Conference Series (Vol. 930, No. 1, p. 012001). IOP Publishing.

Waga, M., Akazaki, T., & Hasuo, I. (2016, August). A boyer-moore type algorithm for timed pattern matching. In International Conference on Formal Modeling and Analysis of Timed Systems (pp. 121-139). Springer, Cham.