EMPLOYEE DATA MINING BASED ON TEXT AND IMAGE PROCESSING

— Employees of any company need to know whether their employees are happy or sad or they have any problem in their working environment. There should be some mechanism to handle this information about the employee. Employee chatting messages could be analyzed using sentiment analysis and employee mood detection is retrieved based on text analysis. Also, Employee facial expressions can be detected using Image Processing on employee images taken through Web Camera while an employee is chatting with colleagues. Using Image Processing, Emotion of employee such as Happy, Angry, Sad or Normal is detected. Employee analysis report is shown to company management to find whether the employee is satisfied with company or employee is facing some problem in the working environment.


B. New Distances Combination for Facial Expression
Recognition from image sequence. Author: Fatima Zahra Salmam, Abdellah Madani, and Mohamed Kissi. This paper illustrate that recognize from image sequence take the first and last frame of the facial expression it represent emotion and neutral state. These methods calculate the distances there are six distances of the eleven points. These points are calculated by support descent method ( SDM).By calculating these points we get the emotion of the face whether the human is happy or sad.

C. Image Data Mining From Finanical Document Based
On Wavelet Feature Author: O.El Badawy, M.R. El-Sakka, K.Hassanein, M. S.Kamel We present a framework for clustering and classifying cheque images according to their payee-line content. The features used in the clustering and classification processes are extracted from the wavelet domain by means of threesholding and counting of wavelet coefficients.The feasibility of this framework is tested on a database of 2620 cheque images. This database consists of cheques from 10 different accounts. Each account is written by a different person.

D. Customer Satisfaction Factor Extraction Method
Using Text Mining Author: Yoko Kobayashi; Kazuhiko Tsuda Specifically, the number of the part-time employees, who were atypical, the just-in-time employees, and the special employees increased. An atypical employment pattern is causing instability, given the case that payment and recognition are not commensurate with work and that the strong high production causes a negative image. It is disturbed by the atypical employment issues at most companies. Therefore, education by an enterprise is needed; this is ideally achieved in the short term through high efficiency.

E. Free Text Mining of TCM Medical records Based On
Conditional Random Fields Author: Qiyu Jiang; Hongyi Li; Jiafen Liang Medical records of Traditional Chinese Medicine (TCM) are usually free text and unstructured data, how to extract medical terms from TCM medical records based on conditional random fields is an interesting problem.TCM medical records obtained from dermatology in Guangdong

Proceedings of 1st Shri Chhatrapati Shivaji Maharaj QIP Conference on Engineering Innovations Organized by Shri. Chhatrapati Shivaji Maharaj College of Engineering, Nepti, Ahmednagar In Association with JournalNX -A Multidisciplinary Peer Reviewed Journal, ISSN No: 2581-4230
21st -22nd February, 2018 Provincial Hospital of Chinese Medicine are segmented to single words and labeled with grammatical properties of words by TCM expert, and divided into training sets and testing sets.

III. PROPOSED SYSTEM
To identify emotion from the facial expression there are three steps they are: face detection, face extraction, classification of the emotions. Firstly we will detect the face by using viola Jones algorithm viola Jones algorithm is used for detecting the faces. Viola Jones is just giving a thousand no of faces to a computer or a digital camera that faces might be know faces or unknown faces to the computer once it get trained to identify whether it is human face or not then it will extract all the features and then classify.
This system is used in maintaining employee data with the help of a database. Live Images from Web Camera and Text Messages will be sent while Employee is chatting with the colleague. With the help of image processing and face recognition algorithm we analyze mood of an employee. In this system we use NLP algorithm for sentiment of messages for emotion analysis. A Report of Employee based on Messages and Mood is generated automatically and shown to Admin of Employee Network. Our proposed approach is inspired from Hammal Calpier work and focuses on calculating six distances presented in Fig,  between four parts of the face. Firstly, forty nine facial characteristic points in below fig. are detected using the SDM Method. Then just eleven characteristic points are considered to calculate six distances, and only the first and the last frames are considered from all sequence frames that represent respectively the neutral state and the emotion state. From these eleven detected points that are represented by x and y coordinates, and refer to the internal parts of the face, we have calculated Euclidian distances. After that, we have measured temporal deformation by calculating the ratio between frames; each calculated distance of the first frame is divided by the same calculated distance of the peak frame to calculate dynamic features.

IV. CONCLUSION
In this system with the help of image processing and face recognition algorithm we analyze mood of an employee. This system will use NLP algorithm for sentiment of messages for emotion analysis. A Report of Employee based on Messages and Mood will be generated. In image processing we use NLP algorithm. By this we will get to known mood of employee text processing the friend will chat with this another friend by that we will known the felling of the employee whether he is happy or not with his work.