MODULE : UU-MBA-714
HEADING: THE FIRST NATIONAL BANK LESOTHO(FNB) USE OF DATA MINING AND KNOWLEDGE MANAGEMENT
TABLE OF CONTENTS
- INTERNAL DISTRIBUTION, PRESENTATION AND MANAGEMENT BUY IN……4
- DATA MINING AND ITS IMPORTANCE ON FNB……………………………………4
- CREATING SALES OPPORTUNITY……………………………………………5
- RETAINING CUSTOMERS………………………………………………………5
- DETECTING FRUAD…………………………………………………………….6
- DATA MINING TECHNIQUES AND ALGORITHMS …………………………………6
- DATA MINING KNOWLEDGE DISCOVERY………………………………………….6
- SOURCES OF DATA …………………………………………………………………….7
- KNOWLEDGE MANAGEMENT IN FNB……………………………………………….8
- HOW FNB CAN BE A LEARNING ORGANISATION………………………………….9
- COMPETITORS’ SIMILAR PRODUCT ANALYSIS………………………………….10
- FINANCIAL AND COST ASSOCIATED …………………………………….……11
- INFLUENCE ON FNB WORKPLACE………………………………………………11
- TABLE OF CONTENTS…………………………………………………………………………………13
In information technology time, information is being very important businesses’ asset that give competitive edge, and this brought an increase in knowledge management techniques. Various companies had gathered and retained a lot of information. Despite all these, they find it difficult to find crucial details of concealed information by converting this information into important and crucial wisdom. Administering knowledge asset is difficult and many companies are incorporating technology in managing their knowledge to assist innovation, transferring, infusion and conveyance of knowledge (Tuamsuk &Silwattananusarn,2012. Knowledge management is a course or measures taken to utilise available data. It based on extracting important information from a large set of data and data mining is critical for knowledge management,( Wang & Wang 2008) argue that mining of data is important in sharing Business Intelligence, common knowledge and utilise data mining as a weapon to enhance people’s knowledge. Data mining reveals the concealed information in data.
In financial sector globally, the primitive brick and mortar banking where the bank and clients physically meet is no more and electronic banking had taken over, cutting expenses and saving time when doing banking driving efficiency. Utilisation of the internet, optimisation of the bank processes and computerised applications had altered the initial notion of banking and the way financial sector operates. From the 90’s, the whole notion of banking sector had changed to internet banking, stored information and auto teller machines globally simplifying banking services and stronger. The data falls among the crucial possession of any banking institution only if they have ability to extract fruitful details from unstructured data. Mining of data enables obtaining knowledge from old data and predicts the future incidents. It assists in the processes of decision making ( Tuamsuk &Silwattananusarn,2012. This paper is going to demonstrate how important it is for First National Bank to mine its date to drive knowledge management to have a competitive advantage. It is one of the top three largest subsidiaries of FirstRand. It is one of the top five largest banks in South Africa. It has many subsidiaries in other countries like Lesotho, South Africa, Swaziland, India, Ghana, Zambia, Mozambique and Botswana. The proposed data mining system’s name is DOMO.
- INTERNAL DISTRIBUTION, PRESENTATION AND MANAGEMENT BUY IN
The document will be communicated following Marx Weber’s a top down bureaucratical model. It will be presented at the board of directors and executive committee for their buy in. The concept will then be presented to the EXCO working committee, Risk committee to analyse and iron out the risk implication associated with the model. We will then look at the regulatory and legal issues with the compliance and legal team. The team will then take the proposal to the Central Bank of Lesotho for approval. When the system is approved, it will go to implementation stage where the middle managers will be convened to be presented this banks innovation as they are the drivers of implementation in various business units. We will then have countrywide roadshows to train staff on the basic uses of the system. Data mining is the major driver of competitive advantage and chances are that management will jump onto this opportunity and invest in it.
- DATA MINING AND ITS IMPORTANCE ON FNB
Data Mining is a crucial move towards Knowledge discovery databases procedures and yields valuable designs from data. This is a process of discovering new trends from abundant collection of data using algorithms to reveal helpful knowledge (Tuamsuk &Silwattananusarn,2012). Data mining is a rational procedure utilised to find in a huge volume of data information that can be helpful. Its aim is to establish prototypes which were not known. When they had been found, they can assist in decision making, development of the strategy and other areas of the business. First National Bank Lesotho has many sources of data that had been gathered and stored over many of its existence in Lesotho which dates to 2004. It had been operating in six out of ten districts in the country with six branches. It has many ATMs and Points of sales machines spread all over the country. FNB is the leading bank in technology through its efficient and reliable online banking and cellphone banking platforms. In Lesotho, it is one of the only four banks that are registered in the country. With all this footprint from countrywide, FNB can create a lot of marketing opportunities.
- CREATING SALES OPPORTUNITY
FNB should devise the means of correctly capturing this data and use it as selling platforms. For existing clients who come into the branch, it should establish each clients’ trend in individual banking needs and use it to cross sell other products related to that, for instance for someone who usually print statements in the branch, they should be offered online banking where they will freely and conveniently do it themselves. This will save FNB stationary and time to assist such clients as they will be using self service channels. For someone who doesn’t have an FNB account but regularly deposit cash to FNB account maybe convinced to open FNB account and transfer funds from his/her own account to the one they regularly deposit to. This will increase the bank’s client base and reduce costs and save time for such individual.
- RETAINING CUSTOMERS
In the modern era, clients have variety of choices of where to bank. Top management in the banking industry should realise that if they don’t focus and take courtesy to each client, the clients can resort to another bank that takes care of them. Mining data assist in new to bank clients focus for banking solutions and revealing the prospect consumption trends for the bank to keep its clients by giving them appreciation tokens that are specifically designed for their banking requirements. Product mix is one of the crucial challenges in the current banking environment. Customer attrition is costly as well as new client acquisition. Prognostic data extraction is helpful to change significant information to knowledge. There are three stages must take place in order to keep clients, gauging the ability to retain clients, digging deeper the root cause of attrition caused by major service failures and coming up with damage control techniques to enhance retention strategy. (Chitra and Subashini,2013)
- DETECTING FRUAD
Fraud is a huge challenge in the banking industry. Sensing and avoiding fraud is not easy as these imposters always come up with different scams which are always becoming complicated to avoid being sensed easily. Defrauding banks is considered as one of the statutory offences in various states, explained as plotting to get assets or funds from any bank or financial institution. It is also known as “ white collar crime”. All the important functional segments of the bank are at risk of being penetrated by fraud with the increasing cases on the liabilities, assets and electronic funds transfers and payments. In the growing states like Lesotho, banks are exposed to this risk. The ability to detect this crime is very minimal. The automated loan endorsements are the biggest considerable method used by banks. Fraud may be avoided by good decision making on authorisation of loan applications by utilising the categorisation models founded on decision trees which are C5.0 A& Cart, Logistic Regression methods and support vector machine. They stop fraud before it takes place. At other times certain part of the population and the historical transaction pattern of clients might plot bank fraud. Mining of data methods assists in learning these trends and banking history that may drive to fraudulent activity. The banks strive to curb fraudulent activities and containment of fraud is knowledge-rigorous action. The model is crucial in fraud sensing of those deceitful banking activities that the client will be performing (Chitra and Subashini,2013).
- DATA MINING TECHNIQUES AND ALGORITHMS
Data mining systems stipulate the differing challenges which can be sculpted and resolved. Data mining can be classified into supervised learning and unsupervised learning notions extracted form machine learning science which is the sub-set of artificial intelligence. Artificial intelligence is described as putting in place and learning systems that shows self- directed acumen or character independently. Machine learning concerns the methods which allows machines to study from the functions they perform and enhance the way the perform. Data mining uses machine learning notions to data. Supervised learning is also regarded as guided learning. The learning procedure is guided by the historical experience, reliant characteristic or aim. It tries to clarify the character of the aim as a role of a set of self-reliant characters or
prognosticators. Directed learning commonly exhibits prognostic models. They differ with unsupervised learning in a sense that examination is done by the software looking at various instances, but the analysts already know what they are looking for. In the learning task, the model study the judgement for forecasting e.g., the model that intends to recognise the clients who tends to react to a advertising should be skilled through examining the behaviour of what were the results that transpired in the past on who did and who did not respond to the initiative. Different algorithms and methods like categorising, gathering, regression, neural networks, decision trees, artificial intelligence, genetic algorithm, nearest neighbour method and many more assist in recovering knowledge from databases.
- DATA MINING KNOWLEDGE DISCOVERY
This discovery is an exercise of selecting a truthful, likely helpful and comprehensible model in data as knowledge assessment or development of new features. Below measures are followed;
- Cleaning of data where unnecessary and unreliable data is removed.
- Incorporation of data which is mixing data from different origins.
- Data assortment is the exercise of extracting needed data from the folders.
- Data conversion is whereby data is altered or combined to a proper format for extraction by precising and totalling.
- Data extraction is the important task of using intelligent techniques to get data models.
- Model Appraisal occurs when models found during data extraction are altered into knowledge based on some intelligent methods.
- Knowledge Reporting is exhibition method that helps in staging knowledge to the people who need it (Rajabhushanam and Yasvand, 2017)
- SOURCES OF DATA
Banking sector is where enormous volumes of data is being generated. Such data is usually coming from the customers credit facility services, investments, insurances and normal transacting. Apparently important insights on the clients’ profile concerning finances is concealed in the huge active databases and these insights may help the banks effectiveness ( Scott et. Al, 2018).FNB have many walk in customers which include FNB banked and non-banked. The banked customers are the existing clients who come into the branch to deposit cash, withdraw cash, pin replacement, card replacement, statements, queries, complaints and many more banking needs while the non FNB banked are those clients who enter the bank solely for depositing cash in FNB clients’ accounts. The second source of data is the system called Hogan system. This is the main system that is used by the bank to capture customer information during onboarding. During the initial inception of the relationship between the bank and the client, we first must gather the customer details which in banking jargon is KYC (Know your customer). This is the mainframe system or the mother of FNB systems. It stores the current FNB active customers base as well as the former customers who once had accounts but now closed. The kind of information that is captured here is customers names, age, marital status, educational level, nationality, employment status, amount of salary, physical address and whether the client stays at their own houses or rented flats. Possession of relevant data does not end there, it leads to our next concept of Knowledge management.
- KNOWLEDGE MANAGEMENT IN FNB
Mclnerney (2002) opine that knowledge management is an attempt to expand valuable wisdom in a firm. In order to promote Knowledge management in an organisation, management must create a conducive environment for interaction, training platforms, and distribution of required information relics. This procedure is directed on information flow and procedural establishment, transferring and dissemination of information. In this regard, knowledge management is a deliberate goal of acquiring the relevant insight to the relevant people at the relevant moment assisting staff to transfer it and implement it in a quest to enhance efficiency (O’Dell & Greyson,1998 & Davenport&Pusuk,1998). It is derived from the existing information that is already stored in a company through effective management of information systems, good response to change and management of personnel derived from good data mining. It is generally explained as a controlled, inclusive method of implementing skills efficiently to gain competitive edge. Arkell, 2007 also indicated that procedures, resources, models and methods which assist the staff member to grasp and transfer what they know successfully. Pearce- Moses further defines it as the management and monitoring of the firm’s intellectual resources through controlling of the insights and which is consumed by the firm to reach its potential limit. It includes the recognition and examination of knowledge possessed and lacking in an organisation that is linked the procedures, strategy and management activities ( Macintosh,1999) as modern banks are now selling variety of products and services that they didn’t sell before such as insurances and mobile phones in an attempt to stay ahead. It can have enough knowledge management if it becomes a learning organisation.
- HOW FNB CAN BE A LEARNING ORGANISATION
Organisational learning is described as a studying initiative by human socialisation as teams or the entire company. When the firm is learning, the entire entity or subsets are adaptable to change in the industry by catching up with the new patterns (Argyris,1999). The implication is that the knowledge that such entity will have will be improved leading to creativity and innovation in such company. It brings required circumstances for the strategic review which doesn’t compromise the going concern and alteration at the company level. Regeneration needs an institution to look at and study fresh approaches but in the same time using the existing information already possessed (Crossan, Lane & White,1999). Selecting, comprehending and controlling knowledge use and examination in a manner which minimize the conflict with both bring organisational learning. Institution regeneration may be visionary if the exercise may include the entire institution, rather that certain teams or people, also the institution should have open door policy. When coming up with this notion of learning organisation, they explained it on four bases;
- Learning organisation shoulders a conflict rising between knowledge usage and finding.
- It is at three stages, one person, a team and the institution.
- There is a relationship at the above three stages of learning which is through mental and team, procedures, perception, understanding, assimilation and institutionalising.
- Thoughts determine activities and activities determines thoughts.
- COMPETITORS’ SIMILAR PRODUCT ANALYSIS
As FNB is one of the four banks in Lesotho namely; Standard Bank, Nedbank and Postbank. It faces the major competition of four banks and two mobile networks as they also provide transactional, money transfers, saving and insurances. There is also big money lender, Letsegho Financial Services whom we compete on lending facilities. Table 1 below shows the competitors similar techniques and how they pose a threat to our system.
|Standard Bank||CORE system||Captures, aggregates and Store customer profile||The system is slow. Overcharges clients|
|Nedbank||Nedbank Homepage Portal||Clients can apply for services online||The don’t have instant online banking transaction|
|Postbank||Mobile app||Clients can bank through the phone and withdraw at retailers||Highly unreliable and usually have network issues.|
|Letsegho Financial services||Smart instalment collection tool||They collect loan instalment from employer, and no one defaults||They give clients too much loans which clients can’t afford and tarnish their image|
|Vodacom Network||MPESA||Clients can access in rural areas with less costs||Have a limited limit and face regulatory issues currently|
|Econet Network||ECOCASH||Clients can access in rural areas with less costs||Not accessible in most rural population|
- FINANCIAL AND COST ASSOCIATED
FNB will have to buy the mentioned software and systems as well as employing or training its own data scientist within its current staff pool. The costs will be as tabulated in table 2 below;
|DOMO SOFTWARES AND SYSTEM||TO PERORM DATA MINING TASKS||$25,000.00 per annum|
|EXTERNAL RECRUITMENT OF DATA SCIENTIST||EXPERT EXECUTION OF DATA MINING||$39,660.00 per annum per individual|
|TRAINING EXISTING POTENTIAL DATA SCIENTIST||TO BE TRAINED BY THE BANK TO FILL THE ROLE||$9000.00 per individual per course|
|TOTAL||$73,660 for the project|
- INFLUENCE ON FNB WORKPLACE
This change will affect FNB in such that there be skills gap, and this will affect its human resources. It will also affect its finances due the investments that will be needed. Information technology will also be involved. It will need well honed, talented data scientist to achieve this competitive advantage. In the data-influence era we are living in, data scientists are a burning concept and highly demanded. The aim is to get the finest data scientist. Currently, analysts believe that millions of positions of data scientists are likely to be not filled because of the scarcity of such individuals at disposal. The worldwide hunt of honed data scientists is not just a hunt for someone with statistics or computer science qualification. They are looking for the knowledgeable persons who have the skills and experience in software programming and analytics together with outstanding presentation skills (Haider, 2015).
We the world of fourth industrial revolution and the quest for companies to minimise costs and maximise profit, it is important for banks to invest in technology. The world we are living in is influenced by information technology at all facets of life. It is high time the banks embrace this new economy. In this race, they should not ignore the fact that technology led to diminishing in personal feel of clients when being assisted. In this case, banks should develop deep knowledge of their clients through data mining and knowledge management. Banks should put in place the systems that will prevent fraudsters from taking advantage. The cost benefit analyses at the glimpse indicate that FNB will get return on investment. It is highly recommended for the management to embrace this proposal.
Argyris, C. (1999). On organizational learning. 2nd Edition. Oxford, UK: Blackwell Business
Arkell, D. (2007, October). Get our heads into it. Frontiers.
Chitra, K., Subashini, B. (2013). Data Mining Techniques and its Applications in Banking Sector. International Journal of Emerging Technology and Advanced Engineering.
Crossan, M.M., Lane, H.W. & White, R.E. (1999). An organizational learning framework: from intuition to institution. Academy of Management Review, 24(3), 532-537.
Davenport, T. H., & Prusak, L. (1998). Working knowledge: how organizations manage what they know. Boston, Mass: Harvard Business School Press.
Haider, M. (2015). Getting Started with Data Science. Publisher: IBM Press; 1 edition
Macintosh, A. (1999). Knowledge Management. Retrieved December 17, 2014, from http://www.aiai.ed.ac.uk/~alm/kamlnks.html
Mclnerney, C. (2002) Knowledge management and dynamic nature of knowledge. Journal for American Society for information science and technology.
Pearce-Moses, R. (2005). A Glossary of Archival and Records Terminology. Chicago, IL: The Society of American Archivists.
Rajabhushanam, C. Yasvand cumaar,S. (2017). Study of data mining in banking sector. International Journal of Pure and Applied Mathematics.
O’Dell, C., & Grayson, C. J. (1998). If only we knew what we know: the transfer of internal knowledge and best practice. New York: Free Press.
Scott, R.I., Svinterikou, S., Tjortjis, C. Keane, A. Experiences of using Data Mining in a Banking Application.
Department of Computation, UMIST.
Silwattananusarn, T. Tuamsuk, K. (2012). Data Mining and its application for Knowledge management: A literature Review from 2007 to 2012. International Journal of data mining & knowledge management process.
Wang, H. & Wang, S. (2008). A knowledge management approach to data mining process for business intelligence. Industrial Management & Data Systems, 108(5), 622-634.